Junichi Toyota, Ian Richards, Borko Kovačević, Marina Shchepetunina - Vision Beyond Visual Perception-Cambridge Scholars Publishing (2017)

Vision beyond Visual
Perception
Vision beyond Visual
Perception
Edited by
Junichi Toyota,
Ian Richards,
Borko Kovačević
and Marina Shchepetunina
Vision beyond Visual Perception
Edited by Junichi Toyota, Ian Richards, Borko Kovačević

and Marina Shchepetunina
This book first published 2017
Cambridge Scholars Publishing
Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
Copyright © 2017 by Junichi Toyota, Ian Richards, Borko Kovačević,

Marina Shchepetunina and contributors
All rights for this book reserved. No part of this book may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, without
the prior permission of the copyright owner.
ISBN (10): 1-4438-8814-1

ISBN (13): 978-1-4438-8814-1
TABLE OF CONTENTS
Acknowledgements ................................................................................... vii
Abbreviations ............................................................................................. ix
Chapter One ................................................................................................. 1

Different Views on Vision
Junichi Toyota and Ian Richards
Part One: Socio-cultural Studies
Chapter Two .............................................................................................. 11

New Zealand’s South Island Myth: The Evolution of a Literary Idea
Ian Richards
Chapter Three ............................................................................................ 25

The Power of Vision in Mythological Thinking: ‘To See the Forbidden’
in Japanese Myths and Fairy Tales
Marina Shchepetunina
Part Two: Cognitive-Semiotic Studies
Chapter Four .............................................................................................. 39

Grammar in the Mind in Relation to Vision: An Analysis of Null-subject
Languages
Junichi Toyota
Chapter Five .............................................................................................. 57

Motivations for Counter Symbolism: A Case of Demonstratives
in Cushitic Languages
Junichi Toyota
Chapter Six ................................................................................................ 69

Comparative Analysis of Ancient Chinese Characters and Modern
Pictograms
Daria Vinogradova
vi Table of Contents
Part Three: Anthropological and Linguistic Studies
Chapter Seven............................................................................................ 83
Vision in African Languages
Marilena Stuwe-Thanasoula
Chapter Eight ........................................................................................... 103

On Expressions of Vision and Other Sensory Perceptions in Serbian
Junichi Toyota and Borko Kovaþeviü
Contributors ............................................................................................. 123
Index ....................................................................................................... 125

ACKNOWLEDGEMENTS
The editors wish to express grateful thanks to Margaret Edgcumbe for

kind permission to quote from the poetry of Kendrick Smithyman in
Chapter 2.
ABBREVIATIONS
1 = first person
3 = third person
1TR = one-argument transitive verb
ACC = accusative
ANA = anaphoric
AP = antipassive
ART = article
ASP = aspect
CAUS = causative
COM = comitative
COMPL = complementiser
COREF = coreferential
DAT = dative
DEF = definite marker
DP = demonstrative pronoun
DP.AUD.NVIS = demonstrative pronoun audible non-visible
EVID = evidential
EXP = experiencer
F = feminine
GEN = genitive
GOAL = goal
IMP = imperative
IMPFV = imperfective
INS = instrumental
IPFV = imperfective
INAN = inanimate
INST = instrument
LOC = locative
M = masculine
MOD = modifier
N = neuter
NEG = negative
NF = non-finite
NOM = nominative
O = object
x Abbreviations
OBJ = object
OBL = oblique
REF = reflexive
PASS = passive
PL = plural
POSS = possessive
PP = prepositional phrase
PRED = predicate
PROG = progressive
PRF = perfect
PRFV = perfective
PRS = present tense
PRT = participle
PST = past tense
REFL = reflexive
REL = relative
REP = reported
Q = question (-formative)
SG = singular
TEMP = temporal
TOP = topic
TR = transitive
CHAPTER ONE
DIFFERENT VIEWS ON VISION
JUNICHI TOYOTA AND IAN RICHARDS
Introduction
Vision plays an essential part in any living creature’s world. Vision is
related to the eyes, and almost all creatures have organs that can be
considered as a receptor of vision. A well-known exception for this is
troglobites which live their entire lives in the dark parts of a cave, and they
normally lack eyes, but have a vestige that used to react to light.
Evolutionary biologists such as Dawkins (1997) and Parker (2004) discuss
the evolution of eyes, and argue that primitive, unicellular organisms, such
as the sea tickle (Noctiluca scintillans), all started to receive a sensation of
light, and eyes started with this function. From this point on, different
creatures evolved their eyes according to their needs, and some retained a
primitive version of ‘eyes’, while others have developed a highly intricate
system, and the high-resolution vision found in human beings, with the
ability to distinguish shapes, distance and colours of objects, is merely one
version of vision. Dawkins assumes that there are at least 40 to 60
different evolutionary paths (Dawkins 1997: 127), but there is one
underlying function, i.e. the receiving of light.
In human culture and cognition, however, the functions of eyes are not
restricted only to the reception of light. Eyes or vision can be used in
various ways, and various metaphorical extensions can suggest that an eye
can be an indicator of the future in many languages, to the extent that it
becomes a grammatical marker (cf. Heine and Kuteva 2002: 128-130).
This is so because eyes are placed on the front side of our body, and if the
future is considered to lie in front of us, eyes are often associated with this
temporal concept and used as a sign to refer to futurity.
Vision in this sense can reflect upon our culture and social history, and
this volume provides an interdisciplinary view on some aspects of vision
in human culture.
2 Chapter One
Linguistics, anthropology and perception so far

Expressions relating to perception have attracted much attention from
interdisciplinary researchers working on languages. Notable studies are
Viberg (1984) and Evans and Wilkinson (2000), who report a case of
languages in Australia and Papua New Guinea where a verb of hearing
plays a major role. Viberg (1984: 136) presents a hierarchical order of
perception, as represented in Figure 1. This hierarchy shows that vision is
the prime perception in humans. This can be shown in the metaphorical
extension of various perceptions, and the vision-based metaphor is perhaps
most commonly found. For instance, earlier verbs of vision developed into
verbs of knowledge in Indo-European languages through metaphorisation,
i.e. a Proto-Indo-European verb *weyd ‘see’ became the English wise or
wit or Irish fíos ‘knowledge’. Similarly, the development of the Proto-
Germanic wáit ‘I know’ originates from the same Proto-Indo-European
verb, but its perfective sense ‘I have completed seeing’ was not shifted to
the past tense ‘I saw/have seen’, but rather to ‘I know’. One may not be
aware that the verb ‘know’ is related to vision in modern languages, e.g.
the German wissen ‘know’, but instances like these show the dominance
of vision in human cognition. This vision-centred perception is
schematically represented in Figure 2, which suggests the applicability of
vision to other perceptions.
Smell
Sight > Hearing > Touch >
Taste
Figure 1. Hierarchical order of perception
HEARING SMELL -contact
SIGHT
TOUCH TASTE +contact
Figure 2. Semantic extensions in perceptual modalities (Viberg 1984: 147)
However, there are other languages in the world, i.e. Australian and
Papuan languages as well as the Bantu languages spoken in East Africa,
where a verb of hearing plays a major role and it is used as a base for a
Different Views on Vision 3
metaphorical extension referring to cognition, e.g. ‘I hear your point’

meaning ‘I understand your point’. For instance, an Australian language
Pitjantjatjara has a verb kulini ‘hear’, as in (1a), and it is highly
polysemous. Among various senses, this verb can be used as a verb of
cognition, as exemplified in (1b). This is not what is expected in, for
instance, Indo-European languages, and what is unique in these languages
is that verbs of hearing seem to be the prime source for semantic
extensions. Evans and Wilkinson (2000) revise Figure 2 as Figure 3. In
Figure 3, it is clear that the verb of hearing plays a central role, and the
dotted line here shows a dubious case and this extension is dependent on
how one interprets data and thus they leave it open for interpretation.
Pitjantjatjara (Australian, Evans and Wilkinson 2000: 563, 564)

(1) a. Ngayulu anangu-ngku wangkanytjala kulinu
I people-ERG talk.NOMZR.LOC hear.PST
‘I hear people talking.’
b. Mutuka/ compyter ngayulu putu kulini
car computer I in.van understand.PRS
‘I don’t understand cars/computers.’
feel
hearing smell
sight
touch taste
Figure 3. Semantic extensions across perceptual modalities in Australian

languages (Evans and Wilkinson 2000: 560)
The diversity here may puzzle some, but one possible interpretation is
found in cultural difference. Sasha Aikhenvald (p.c.) suggests that those
languages that put emphasis on the verb of hearing are spoken in a culture
where religiously-gifted people (i.e. shamans, spiritual healers, etc.) have a
special power and a social role in their local society. They are said to be
able to see ‘everything’, including ancestral spirits. Thus, the verb of
vision is reserved to these gifted people and the common people have
resorted to the second most prominent perception, i.e. hearing, and this is
how the verb of hearing became prominent in languages in specific parts
of the world.
4 Chapter One
A similar interesting case related to local religious beliefs is found in

East Africa, where an olfactory verb can be a base for semantic extension.
Thanassoula (2013) reports the case of Lussesse, a Bantu language spoken
on the Ssesse Island in Lake Victoria. This language also has a highly
polysemous verb of hearing -húlirà, similar to Australian and Papuan
languages. This suggests a close link between the use of perception verbs
and local religions. What is peculiar in East Africa is that people believe
that ancestors communicate through smell according to a local religion in
the region, and only religiously-gifted people can interpret smells. In this
culture, smell gains a special status among different perceptions, unlike
anywhere else in the world. As Figure 4 shows, -núuka ‘smell good’ can
be extended to cognition, and this use is only found among religious
people, and common people use the verb of hearing for various extensions
including cognition. The use of an olfactory verb as a base for semantic
extension is typologically rare, but the case in East Africa reinforces a link
between religious influence and use of perception verbs as suggested by
Aikhenvald.
Figure 4. Lussesse (Bantu) perception (Thanassoula 2013: 255)
As shown here, previous research on perception is normally in

relation to psychology/cognition or anthropology. Perception is indeed a
multidimensional topic, but purely linguistic analysis dealing with internal

structures of perception has rarely been done, except for a description of
structures in reference grammars. In what follows, internal linguistic
structures denoting perception will be analysed, stemming from an
analysis of a system in Serbian.
Literary studies and reflection on culture

It can be argued that a major part of the Modern movement in western
culture is the recognition of the slippery relationship between perception
and its representation, whether in speech or in art. Visual perception has
been a factor in this almost from the beginning, such as when the painter
Claude Monet surprised the public in 1874 with his painting ‘Impression,
Sunrise’, a painting that did not attempt to offer up a perfect copy of what
the eye might see, but rather gave an indication of what the mind might
register as significant in the data passed to it by the perceiving eye. A little
later the linguist Ferdinand de Saussure was explaining that language is at
best an arbitrary collection of signs agreed on by a community, and the
poet Wallace Stevens was showing in ‘Thirteen Ways of Looking at a
Blackbird’ that art could behold but never contain reality, since a
fourteenth way of looking at a blackbird would always exist.
Perception has a strong element of cultural construction, since the
brain must organize what it senses, particularly its visual information, and
this is then mediated through language. Modern art and letters have always
been acutely conscious of this fact, and a perfect example of what happens
when the process breaks down is supplied by Jean-Paul Sartre in his
classic existentialist novel, Nausea, as the hero, Roquentin, perceives the
black root of a chestnut tree in a park without the shaping benefit of the
labels supplied by language.
I was thinking without words, about things, with things […] That black
there, against my foot, didn’t look like black, but rather the confused effort
to imagine black by somebody who had never seen black and who
wouldn’t have known how to stop, who would have imagined an
ambiguous creature beyond the colours. It resembled a colour but also…a
bruise or again a secretion, a yolk—and something else, a smell for
example, it melted into a smell of wet earth, of warm, moist wood, into a
black smell spread like varnish over that sinewy wood, into a taste of sweet,
pulped fibre. (Sartre 1938: 185-7)
6 Chapter One
Significantly, Sartre begins with a visual image in his efforts to break

down the reader’s (and his hero’s) certainties of perception, since it is our
visual sense which is dominant, and of which we feel the surest.
The idea of having never seen the colour black may seem strange to us
at first—although it has been demonstrated that different cultures have
widely different terms for their perceptions of the colour spectrum—but
the issue of perceiving, labelling and understanding has been highlighted
in western culture by the voyages of exploration that followed Christopher
Columbus’s discovery of the Americas in the late fifteenth century.
Explorers, and then settlers, had to come to terms with new environments
for which their language and art forms could not easily account: these
forms were then forced to adapt. This adaption process has led directly to
what is now termed Post-Colonial literature and art, and the case of New
Zealand literature is featured in this volume as an example.
But problems of adaption can be a two-way process, and just as new
environments can cause difficulties with perception and representation, so
too the relationship between old-world and new-world cultures can create
difficulties of perception for those who have been raised in new-world
cultures. The Caribbean writer V.S. Naipaul focuses on this peculiarity in
his essay ‘Jasmine’. He writes of how, having grown up in Trinidad, he
was familiar as a schoolboy with the works of British literature long
before he had any concrete experience, visual or otherwise, of what he was
reading. This fact tended to turn the English language into a kind of game
in which words existed without referents, and occasionally vice-versa, a
game where his old-world cultural heritage could interfere even with his
perceptions of his own environment.
A little over three years ago I was in British Guiana. […] Suddenly the
tropical daylight was gone, and from the garden came the scent of a flower.
I knew the flower from my childhood; yet I had never found out its name. I
asked now.
‘We call it jasmine.’
Jasmine! So I had known it all those years! To me it had been a word
in a book, a word to play with, something removed from the dull
vegetation I knew. […] Jasmine. Jasmine. But the word and the flower had
been separate in my mind for too long. They did not come together.
(Naipaul 1972: 30-1)
As shown here, cultural issues will always complicate perception of any

kind as the brain tries to distinguish and order its information, but perhaps
these issues will always affect visual perception most of all, since it is our
surest source of contact with our world. Certainly it is one of the roles of
the creative artist to defamiliarize the world for us and make us see with
fresh eyes what is before us, and also to remind us of the extraordinary degree
of subjectivity with which we all perceive our common environment.
Topics covered in this volume

This volume consists of three parts, i.e. socio-cultural studies, cognitive-
semiotic studies, anthropological and linguistic studies. The first part,
socio-cultural studies, contains two papers. The first paper by Richards
discusses the struggles of earlier writers in New Zealand with landscape,
known as the South Island Myth, and presents how vision played a role in
establishing a base in a formerly Philistine New Zealand for its future
literature. The second paper by Shchepetunina analyses the meanings of
vision in various mythological stories, particularly African, Ancient Greek
and Ancient Japanese myths. A particular focus is made on the forbidden
gaze which may lead to creation in some cases.
The second part focuses on cognitive-semiotic aspects of vision.
Toyota contributes two papers here: one is about languages without an
overtly expressed grammatical subject, known as null-subject languages,
and how internal vision can aid the grammar to augment a subject referent.
His other paper deals with exceptional cases of sound symbolism focusing
on a Cushitic language, Somali, arguing that what is referred to by sound
symbolism may not be restricted to actual size, but includes a size
perceived in one’s vision in relation to distance. Vinogradova analyses
pictograms, including the modern emoji characters and ancient Chinese
characters, paying attention to the position of components and its influence
on meanings, and the modification of meanings, as well as the
interpretation of movements in a static representation.
The third part, anthropological and linguistic studies, contains two
papers dealing with linguistic aspects of vision. Stuwe-Thanasoula
presents current research on perception, especially vision, in African
languages, comparing various languages in different language families and
regions in Africa. Toyota and Kovaþeviü work on a case study of
perception in Serbian, presenting how linguistically creative Serbian as
well as other Slavic languages can be when it comes to expressions of
perception, including vision.
8 Chapter One
References
Dawkins, R. 1997. Climbing mount improbable. London: Penguin Books.
Evans, N. & D. Wilkins 2000. In the mind’s ear: the semantic extensions
of perception verbs in Australian languages. Language, 76, 546-591.
Naipaul, V.S. 1972. The Overcrowded Barracoon. Harmondsworth:
Penguin Books.
Parker, Andrew 2004. In the blink of an eye: how vision sparked the big
bang of evolution. New York: Basic Books.
Sartre, J.-P. 1938. Nausée. Translated by R. Baldick 1965. Nausea.
Harmondsworth: Penguin Books.
Thanassoula, M. 2013. Perception in Lussesee (Bantu, J10). In A.
Aikhenvald & A. Storch (Eds.), Perception and Cognition in Language
and Culture, (pp. 251-270), Leiden: Brill.
Viberg, Å. 1984. ‘The verbs of perception: a typological study.’ In B.
Butterworth, B. Comrie and Ö. Dahl (Eds.) Explanations for Language
Universals, (pp. 123-162). Berlin: Mouton de Gryuter.
PART ONE:
SOCIO-CULTURAL STUDIES
CHAPTER TWO
NEW ZEALAND’S SOUTH ISLAND MYTH:

THE EVOLUTION OF A LITERARY IDEA
IAN RICHARDS
Abstract. On their arrival in New Zealand from Europe, the first settlers
felt a profound sense of alienation from their adopted environment and a
resulting sense of unease about their place within it. This view coalesced
into an idea known as the ‘South Island Myth’, a view that the landscape
remained coldly indifferent and even hostile to European settlement. This
indifference could thus often be damaging to the settlers, blighting their
physical and mental lives. The South Island Myth then permeated early
New Zealand writing, especially the country’s poetry, and still later
transformed itself into a more provincial complaint by writers about the
difficulties of pursuing the arts and culture in a largely Philistine land.
Finally with the nation’s development the South Island Myth began to
recede from the local literary scene, but it was then playfully revived by
Post-Modernist writers as an idea that could be exploited in poems of wit.
Examples from the works of Edward Tregear, Charles Brasch, Kendrick
Smithyman, and Bill Manhire are used to illustrate this thesis.
Introduction
In 1961 the Polish writer Stanislaw Lem published Solaris, a science-
fiction novel about the discovery by humans of sentient life on a distant
planet. The planet is studied, and scientists soon find to their amazement
that it is the entire planet itself which is alive and not simply some
creatures inhabiting the surface. Attempts are then made to observe and
understand this phenomenon, and several chapters in the book detail
catalogues of scientific data concerning the oceans which seem to cover
the planet, though none of this information offers any insight into the
deeper workings of the living organism. The ocean, which at first is
imagined to be a source of the planet’s life, turns out to be only something
else which resembles an ocean. Eventually a space station is built around
12 Chapter Two
the planet that scientists can visit, and in an attempt to communicate with
this baffling life form X-rays are fired into its surface. This action results
in a series of bizarre appearances on the space station of people created by
the planet from the scientists’ own minds, soon revealing a lot about the
scientists themselves but failing to reveal anything about the planet’s
consciousness, much less any way of communicating with it. At the end of
the novel the scientists are no nearer understanding or communicating
with the exotic planetary form of life which they have found than they
were at the beginning.
It is possible to read Solaris as a contemporary updating of the many
European voyages of discovery in the centuries between Christopher
Columbus and James Cook. European explorers, and then settlers, arrived
in lands so alien as to baffle them, and any indigenous inhabitants, to the
extent that these people were regarded as anything other than a hindrance,
were often viewed simply as extensions or reflections of the Europeans
themselves. Bafflement and a limited viewpoint became the natural
responses in an alien world. The scientists in Solaris can understand the
living planet only in terms of its not being like themselves, and the same
was true for European explorers. During exploration and even after
settlement, the landscape—the primary node of contact with the new
environment, a something else which only spuriously resembled things at
home—could offer no easy point of entry. Sometimes a crisis of
perception followed and became a source of anxiety.
For New Zealanders this anxiety, an intense desire to understand and
feel at home in a foreign landscape that they had chosen to inhabit, or even
more alarmingly a still-pristine landscape they had been born and raised in,
permeated through settler life and art from near the very beginning, and
continued to be a common theme among local writers for most of the first
half of the twentieth century. Edward Tregear’s ‘Te Whetu Plains’ is a
case in point. It was published in a collection in 1919, though possibly it
was written as early as 1872. The poem therefore often appears near the
start of anthologies of New Zealand literature since, as Tregear’s
biographer, K.R. Howe, has noted, ‘Te Whetu Plains’ ‘has often been seen
as encapsulating a more general state of mind—that of an immigrant’s
alienation in the strange landscape of a new country’ (Howe 1987: 55). In
the poem Tregear regards Te Whetu Plains at night, an empty place in a
densely forested area—a place notable above all for being dark and
silent—as a kind of early isolation tank. He makes the place stand for an
image, redolent with nineteenth-century religious doubt, of the horror in
any possible life that might exist after death. If the newly scientific view
of the creation of the world is correct and any afterlife does not involve a
New Zealand’s South Island Myth: The Evolution of a Literary Idea 13
conventional Christian heaven, Tregear reasons, then the afterlife would

be somewhere permanently removed from all the sights, sounds, and
sensations of home. It does not matter to Tregear that the plains are in
‘moonlit darkness’; he cannot see anything familiar in the outlines of the
‘giant terraces’ below him. Despite the plentiful bird sounds to be heard
during the New Zealand daytime, at night the antipodean plains are silent,
without birdsong and with no noise from a ‘far-off’ stream, but Tregear
can express this only with a kind of colonial negative capability, as lacking
English and Romantic-literary sounds:
‘tis a songless land

That hears no music of the nightingale,
No sound of waters falling lone and grand
Through sighing forest to the lower vale,
No whisper in the grass, so wan and grey, and pale.
(Wedde & McQueen 1985: 97)
Unlike most birds, nightingales sing after dusk, and moreover, they were a
common Romantic-era trope for the nature-inspired poet, so that Tregear’s
‘songless land’ suggests a cultural as well as a geographic failure. Tregear
may even have in mind John Keats’s Romantic reverie in his famous ‘Ode
to a Nightingale’ as an implicit contrast, where the sound of a nightingale’s
song, allied to ‘the wings of Poesy’, puts the poet into a blissful reverie, a
state which he then compares to the ecstasy of being in heaven after death.
But this New Zealand landscape offers only a ‘ghastly peace’ of silence, a
nothingness for an alienated consciousness totally deprived of any familiar
comforts. Unlike Keats’s cosy garden with its English nightingale singing
in the twilight, the exotic Te Whetu Plains fail as a site for any ecstatic
release from the self in nature. Furthermore, once Tregear feels deprived
by science of the spiritual comforts of Christianity, this landscape no
longer allows him the worship of nature as a substitute for religion.
The 1930s and 1940s

The uneasy relationship to the landscape expressed by Tregear, in which
the environment is viewed wholly in terms of non-recognition and
negatives, so that its birds are not nightingales and its rivers not well-
known streams, coalesced in the later work of several New Zealand-born
writers, most of them based in the South Island, into an attitude, a local
myth, which was then available for poetry. It was arguably the first
authentic literary idea that evolved directly from Pakeha (meaning New
14 Chapter Two
Zealand European) responses to the new landscape. 1 The New Zealand

environment was soon seen as familiar—since it was the only landscape
available for direct experience—but one still unwelcoming and indifferent
to its new inhabitants, unlike the British landscapes of the settlers’ origins.
The land was presented as beautiful but coldly indifferent to settlement. It
was seen as unwelcoming in its rugged contrast to Europe. Sometimes it
was viewed as downright hostile, the sort of place in which people’s
development was stunted and damaged. This unwelcoming quality in the
local landscape appeared to a greater or lesser extent in the work of New
Zealand poets as otherwise diverse as D’Arcy Cresswell, Ursula Bethell,
Charles Brasch, Denis Glover, and Allen Curnow, and it became known
loosely as the ‘South Island Myth’. The critic Lawrence Jones’s excellent
account of the growth of this idea and of the literary movements which
developed in the early 1930s and 40s details how the South Island Myth
grew into fruition as an ‘anti-myth’ to the rosy-tinted view that ‘New
Zealand was “God’s own country”, a pastoral paradise and a Just City,
based on an ideal English model, being perfected through an historical
process of triumphant progress’ (Jones 2003: 173). The South Island Myth
was nicely epitomised by Allen Curnow’s 1941 poem ‘House and Land’
with its Auden-influenced, proletarian-style rhymes describing a New
Zealand ‘spirit of exile’. Its much-quoted closing lines are:
Awareness of what great gloom

Stands in a land of settlers
With never a soul at home. (Wedde & McQueen 1985: 198)
The South Island Myth was also described at length, and often in
mystical terms, in the essays of M.H. Holcroft, and perhaps reached its
high-water mark when Holcroft’s book The Deepening Stream won the
essay prize in the 1940 New Zealand Centennial Literary Competition.
The idea received possibly its most plaintive, Romantic articulation in
Charles Brasch’s poems, notably ‘The Silent Land’, where ‘The plains are
nameless and the cities cry out for meaning’, and where the problem with
the landscape is explicitly diagnosed as a lack of satisfactory history:
Man must lie with the gaunt hills like a lover,

Earning their intimacy in the calm sigh
Of a century of quiet and assiduity (Curnow 1945: 133)
1
The word ‘Pakeha’ is a Maori word, commonly used by all New Zealanders, for
non-indigenous New Zealand people.
It is worth noting that the South Island Myth was, above all, a Pakeha
notion of the land. The Polynesian Maori, the indigenous inhabitants of
New Zealand, featured in this view of the land scarcely at all, a fact that
seems incredible today. But at the time the prevailing colonial view was
that the Maori were a dying race, a tragic group of Romantically dusky
forebears being swept away by more advanced Pakeha arrivals. Furthermore,
those Maori who might survive this replacement process would do so by
becoming thoroughly assimilated into Pakeha ways, so that any remnants
of indigenous familiarity they might have with the landscape would
become, at the very least, irrelevant and quaint. It was a convenient view,
if you were Pakeha, and perhaps it was no coincidence that the South
Island Myth flourished in the South Island, where picture postcard scenery
was plentiful and the Maori population was low. But excluding Maori
from any consideration of the relationship between Pakeha and the new
land also masked a subtler and perhaps darker factor. On the whole, the
alienation of Pakeha from their new environment was seen in the culture
of the South Island Myth as a failure on the part of the environment itself,
and not so much a failure by Pakeha settlers. The Maori, with their own
indigenous culture already adapted to their homeland, could figure in such
a relationship only as a reproach to Pakeha failure or, at best, as an
encouragement, a pointer towards future adaptation by Pakeha and the
possibility of success. But neither of these would foster the notion of
Pakeha as more advanced, civilising arrivals. While the new scenery was
being observed with anxiety, suspicion, or disgust, Maori were best left
out of the picture.
The 1950s and 1960s

Critical accounts of the South Island Myth usually end in the 1930s or 40s,
but the idea itself seems to have hung on and to have been adapted in the
work of later writers. By the 1950s and 60s the prevailing anxiety with the
landscape had transformed itself somewhat among the literati into a kind
of urban hauteur directed towards a rural country: it had become the
complaint of the sensitive artist, an inheritor of European sophistications,
about living in a provincial land which did not much value the arts. This
anxiety no longer indicated solidarity with Pakeha living in the harsh wilds
of a daunting country, but rather it had become an unease about living and
creating art in a hostile land of all-too-happy Philistines, a land where
practical activities counted and where poetry and culture did not. A good
example of this view, and certainly an example of a poem directed at the
16 Chapter Two
urban cognoscenti, is Kendrick Smithyman’s ‘Colville’, which was written

in 1968.
Born in 1922, half a generation after Brasch and Curnow, and having
grown up in Northland and Auckland instead of in the south, Kendrick
Smithyman saw the South Island Myth and its conventions for writing
about the New Zealand landscape from some critical distance, both in
terms of time and geography. Early on, together with fellow aspiring poets
Keith Sinclair and Robert Chapman, he even formed the ‘Mud Flats
School’ in specific opposition to the dewy-eyed mysticism of South-
Island-Myth-inspired writing. The School did not last, but from the outset
Smithyman clearly intended to write poetry that was tougher and smarter
than the work of his predecessors, and which would be more of a direct
response to the New Zealand landscape than a mere complaint of what the
landscape was not. In Smithyman’s papers ‘Colville’ is recorded as having
being written on 11 January 1968, so that the poem may have resulted
from a summer holiday visit, and Smithyman in fact had a son who lived
for a time on a commune in the region (Smithyman 1989: 75). Colville is a
small town in the North Island, located near the top of the Coromandel
peninsula, with its lone general store regarded as a last stop for provisions
and petrol. The area is well known for fishing, and today Colville has an
alternative-lifestyle, cosmopolitan atmosphere that it did not have at the
time Smithyman had in mind when he wrote of it. For this reason
Smithyman at one stage titled the poem ‘Colville 1964’, to place it in the
context of the recent past, although the date was dropped from the title in
his Selected Poems and in later works.2 In its entirety the poem reads:
That sort of place where you stop

long enough to fill the tank, buy plums,
perhaps, and an icecream thing on a stick
while somebody local comes
in, leans on the counter, takes a good look
but does not like what he sees of you,
intangible as menace,
a monotone with a name, as place
it is an aspect of human spirit
(by which shaped), mean, wind-worn. Face
outwards, over the saltings: with what merit
the bay, wise as contrition, shallow
2
Smithyman confirmed all this in a private letter to me written on 17 May, 1994.
as their hold on small repute,

good for dragging nets which men are doing
through channels, disproportioned in the blaze
of hot afternoon’s down-going
to a far, fire-hard tide’s rise
upon the vague where time is distance?
It could be plainly simple

pleasure, but these have another tone
or quality, something aboriginal,
reductive as soil itself—bone
must get close here, final
yet unrefined at all. They endure.
A school, a War Memorial

Hall, the store, neighbourhood of salt
and hills. The road goes through to somewhere else.
Not a geologic fault
line only scars textures of experience.
Defined, plotted; which maps do not speak. (Smithyman 1989: 75)
The poem begins in a dismissive tone, describing the town of Colville

as typical of ‘That sort of place’ where you do not stop long except for
petrol, local products like plums, and an ice-cream snack. The poet
employs an impersonal ‘you’ in speaking which appears to be the product
of his train of thought as events unfold, but which also usefully includes
the reader. The visiting poet’s shopping activities involve entering the
general store and thus, inevitably, being sized up by ‘somebody local’, a
person who comes in and ‘does not like what he sees of you’. This local is
presented as typical, an inarticulate ‘monotone with a name’, someone
hard to read but nevertheless vaguely threatening and therefore ‘intangible
as menace’. The visiting poet concludes that Colville as a place reflects,
and is shaped by, ‘an aspect of human spirit’ which is ‘mean, wind-worn’.
The local environment has damaged the local people.
Turning away from the counter in the store, the poet faces out towards
the view ‘over the saltings’, the narrow bay of coastal water visible nearby,
while he also takes in a broader view of the place as a whole. He asks
himself what ‘merit’ the bay might have. His answer appears in two
gnomic phrases: ‘wise as contrition, shallow/ as their hold on small repute’.
These two lines might be construed as the bay being: ‘wise in the form of
showing some sort of contrition or sorrow for its own sins’ and ‘shallow in
its waters in the same shallow way as the locals have any small claim to a
good reputation’; yet ultimately the extreme compression of the phrases
18 Chapter Two
aims to yield the poet’s fleeting impressions rather than his carefully
formulated thoughts. After these moments of largely abstract speculation
the poet’s mind directs itself further outwards again to take in more
concrete details. He notes the bay’s utility—it may be a handy spot for
fishing with nets—and he observes some men doing just this, ‘dragging
nets’ in the late afternoon heat against the indistinct horizon. Where, the
poet seems to be wondering, does a merely human settlement, someplace
provisional like Colville, really fit into all of this natural vastness and
eternity?
In the next stanza the poet then goes on to reconsider his previous
views.
It could be plainly simple

pleasure, but these have another tone
or quality, something aboriginal,
reductive as soil itself—bone
must get close here, final
yet unrefined at all. They endure. (Smithyman 1989: 75)
He begins by thinking that the men dragging nets might be doing so for
‘plainly simple/ pleasure’, like holidaymakers, but decides instead that
their activities ‘have another tone/ or quality, something aboriginal’. They
are working in the manner of people who have become, at least in part,
indigenous to their environment, and so their fishing is as ‘reductive’, or
simple and unembellished, as the ‘soil’ or land which they live on. The
poet now suddenly muses in more general terms, admonishing himself and
perhaps the reader, that ‘bone/ must get close here’. Just as the beginning
of the stanza played with the expression ‘plain and simple’, rendering it as
‘plainly simple’, so too here the poet plays cleverly with the expression
‘close to the bone’. His cleverness may be a small, linguistic act of
avoidance of the full implications of his thoughts: it is certainly a bit close
to the bone, or true to the point of discomfort, for the poet to acknowledge
that this, the act of dragging nets in a shallow bay, is what it is to be local,
adapted, and like a native, rather than displaying the urban sophistication
which the poet evidenced at the start of the poem and which he has
maintained until this point. Practical work is how you get your living
bones close to the soil. It is the ‘final’ state in the process of adaptation to
the new environment, and yet it is not refined at all. Culture, in the refined
European sense of the word, is not anywhere involved in this process.
‘They endure’, the poet concludes of the local men going about their
fishing, with the weighty implication unspoken that the poet and his big-
city ways, which are merely imported from overseas, will not endure. The
poet’s consideration of the bay, ‘wise as contrition, shallow/ as their hold

on small repute’, which was both unfocused and over-compressed,
suddenly might apply just as well to the poet himself.
Before leaving, the poet has a last look at Colville, taking it in at a
glance. What he sees are buildings related to the simple necessities and
rituals of a settlement, ‘A school, a War Memorial/ Hall, the store’. After
that there are only the nearby hills. The poet announces that ‘The road
goes through to somewhere else’, a line that could have been lifted out of
almost any poem expounding the South Island Myth, except that in this
case the poet’s departure suggests a revived self-importance which is also
perhaps tinged with a sense of relief at his escape. He begins the second
half of his final stanza with the observation that the place he has visited is
not a ‘geologic fault/ line’, or some kind of mistake in the landscape,
(though like most of New Zealand, Colville probably lies along earthquake
fault lines, and the place does perhaps form a fault line, a pressure point, in
the poet’s thinking, albeit one which he is keen to deny). Colville is just a
place which makes ‘scars’ of habitation, the poet decides, as the locals’
experience accumulates of and in this new land. In the last line of the
poem, as the town recedes into the distance for the departing poet and
becomes no more than a reference on his map, Smithyman uses the
ambiguities available in the language of map-reading to note that Colville
can be ‘Defined’ and ‘plotted’ on a map, but the town is defining itself and
plotting its own story in a way which maps, the products of urbanized and
sophisticated methodologies, cannot explain: ‘Defined, plotted; which
maps do not speak.’
Other notable poets writing in a similar vein in the 1950s and 60s are
Peter Bland, Louis Johnson, and even James K. Baxter. Johnson and
Baxter both use images lifted from the landscape to diagnose the
privations that they see as specific to Philistine New Zealand life.
Johnson’s 1952 poem ‘Magpies and Pines’ employs magpies as
threatening birds which have been known to disrupt people’s well-fed
morning reveries and ‘drop through mists of bacon-fat/ with a gleaming
eye, to the road where a child stood screaming’. Johnson also employs ‘the
secretive trees’ in the ‘dark park’, which have to be resorted to by lovers
for stolen kisses or any other sexual activity, to help the poet indict the
puritan repressiveness which blights the lives of New Zealand’s sensitive
young (Wedde & McQueen 1985: 312). Magpies and pine trees are both
introduced species which seem to have displaced what was in the past the
simply natural and pleasurable. As a result, in such a degraded world sex
becomes a complicated and subversive act which finds ‘small truth/ in the
broken silence’. Similarly Baxter’s famous sequence ‘Pig Island Letters’,
20 Chapter Two
published in 1966, uses its opening stanzas to describe an impoverished

landscape, ‘an old house shaded with macrocarpas’, from which the poet
diagnoses a land suffering from a lack of human love and affection
(Wedde & McQueen 1985: 337-8). Baxter claims he is able to diagnose
this suffering accurately because the same lack of love in the land has
affected himself and become ‘my malady’. He suggests that a sense of
alienation is something innate which he shares with his fellow New
Zealanders, so that he ‘will lie some day with their dead.’
Thus it might almost be possible to graph the responses of the first
New Zealand writers to their new homeland, where the vertical axis is an
ascending sense of alienation and the horizontal access is the passage of
time away from first contact. The resulting graph would probably appear
as something like a power-law (or Pareto) graph, plunging steeply from
high up on the vertical axis into a tight turn and then drifting on in a low,
drawn-out trajectory. At the beginning of the graph’s time axis early
writers, like Tregear (anticipating the scientists in Solaris at first contact),
would see their environment wholly in terms of non-recognition and
negatives, that its birds were not nightingales, its rivers not well-known
streams and so on, so that their sense of alienation would rate extremely
high. This sense would diminish with the writers’ greater familiarity with
the environment, exhibiting itself in poems like Allen Curnow’s 1941
‘House and Land’, where the landscape is more familiar but seen as still
unwelcoming. At last it would descend further into poems which present
the New Zealand landscape rather more as somewhere readers would
recognise and feel at home in, but which is still hostile to the arts and to
the activities of artists who would like to transform or influence the local
scene. It would ultimately be impossible to put meaningful labels on these
two axes of the graph, but poems might be placed in an approximate
fashion on the graph’s curve by considering the texts in relation to each
other, and the poems might be expected to appear along the horizontal
time axis more or less in the chronological order of their publication.
Beyond the 1960s

As New Zealand continued to develop and diversify, the essentially
provincial complaints of local writers concerning Philistinism began in
themselves to seem old-fashioned, and the last remnants of the South
Island Myth largely disappeared from the country’s literature. This was
not entirely an end to the matter, however. Once the South Island Myth
had degenerated into little more than an old-fashioned attitude, it became a
trope which revisionist poets could then take advantage of. By the 1980s
New Zealand landscape poetry had moved on to other themes, usually

expressing more personal views, so that even the consideration of the
inevitable onset of death, which appears often in Allen Curnow’s late
‘Kauri Road’ poems, occurs against a scenic backdrop that is no longer
threatening or indifferent and that offers instead imagery which can seem
comforting or familiar. But the South Island myth continued to exist as an
idea that could be updated and played with, which is what the Post-
Modern poet Bill Manhire, born in the South Island in 1946, does in his
well-known poem ‘Zoetropes’. It could be argued that as a poet Manhire
has little interest in contributing any new ideas to New Zealand literature,
but he is certainly adept at using the ideas of his predecessors in clever and
witty ways. Indeed, it is frequently the conventional nature of his topics
that makes it possible for some of his otherwise puzzling poems to yield
up their meanings.
‘Zoetropes’ depicts a state of mind which both is, and is not, entirely
Manhire’s own. It is anchored in time and space with its coda, ‘London
29.4.81’, but the poem describes an experience common to almost all New
Zealanders overseas (Wedde & McQueen 1985: 480). On reading
newspapers, or similar, an expatriated New Zealander’s peripheral vision
tends to react to the unusual capital letter Z, having learned unconsciously
that this will likely refer to news of his own country. This scanning is, to
some degree, a symptom of homesickness. Manhire’s poem, faithful to the
precepts of Symbolism which have influenced him, tries not merely to
describe this experience but to reproduce it. It begins, cheekily using the
letter A, with ‘A starting’. The poet-speaker’s eye jumps to the word he
sees beginning with Z only to be disappointed; he then reads further
among ‘other disappointments’, which are not news of home either. The
poem itself even consists of three separated groups of stanza-pairs that
seem to straggle disjointedly down the page. Indeed, reading a foreign
newspaper can be an unpleasant reminder for New Zealanders of just how
unimportant their little nation is in the world. It is a glimpse, as if through
the slits of a revolving disk, or zoetrope, not of an early version of moving
pictures but of the nothingness of New Zealand on the global stage.
While reading, it seems the speaker’s eye has mistakenly snagged on
‘Zenana’ which, as someone like an impressively literate poet might
explain, is the place in the East where a harem is traditionally hidden from
sight. Something privately valuable and yet not publicly valued, kept out
of sight—this is, in fact, not a bad image for a New Zealander’s view of
his homeland when overseas. But to the extent that the poet-speaker muses
on this at all, he thinks only of another word beginning with Z, ‘Zero’. For
zero, too, is nothing, like the country that is not there. But, just like one’s
22 Chapter Two
remembered homeland when overseas and sizing up the wider world, this
nothingness is also ‘the quiet starting point/ of any scale of measurement’.
The word ‘scale’ then becomes the cue for the poet-speaker’s final,
somewhat plaintive, home thoughts.
The land itself is only

smoke at anchor, drifting above
Antarctica’s white flower,
tied by a thin red line
(5,000 miles) to Valparaiso.
New Zealand’s Maori name, Aotearoa, is usually translated as ‘The land

of the long white cloud,’ and the land does look, on a map, like little more
than a wisp of smoke in the bottom corner, uncomfortably close to a
bulbous Antarctica. In Manhire’s poem the once common concept of New
Zealand as a land defined by its distance from other countries is reduced
solely to noticing a longitudinal marker. The country is 5,000 miles from a
place in Chile which few have ever heard of, ‘tied’ only to further
insignificance. The poem’s throwaway last line seems especially fitting in
this context.
Conclusions
Though a very young country and settled mostly in the nineteenth century,
New Zealand soon raised problems of perception to be faced by early
explorers and settlers. The resulting anxiety, as the critic Alex Calder has
noted, ‘unsettled’ New Zealand poets and writers when they first began to
pursue their art in a new environment and led to the articulation of the so-
called South Island Myth (Calder 1998: 165). But by facing their
environment honestly, especially in terms of their own failure to perceive
anything familiar or friendly in it, budding New Zealand poets were able
to create poems which expressed a purely local experience. The South
Island Myth transformed itself and became something much more like a
form of urban complaint about provincialism as the twentieth century
progressed, albeit a complaint that could be expressed in very
sophisticated poetic forms, and eventually it declined into a mere attitude
which could be subjected to play by Post-Modern writers. However, the
very Pakeha-centred nature of the South Island Myth itself has now
become largely redundant in a country that has embraced multi-
culturalism, and new myths and anti-myths will no doubt follow along the
trail which has been created.
References
Calder, A. 1998. Unsettling Settlement: Poetry and Nationalism in
Aotearoa/New Zealand. In Thomas Brook (Ed.), REAL: Yearbook of
Research in English and American Literature 14, Literature and the
Nation, (pp.165-81). Tubingen: Gunter Narr Verlag.
Curnow, A. 1945. A Book of New Zealand Verse 1923-45. Christchurch:
Caxton Press.
Howe, K.R. 1987. The Dating of Edward Tregear’s ‘Te Whetu Plains’ and
an Unpublished Companion Poem. Journal of New Zealand Literature,
5, 55-60.
Jones, L. 2003. Picking Up the Traces: The Making of a New Zealand
Literary Culture 1932-1945. Wellington: Victoria University Press.
Lem, S. 1961. Solaris. Warsaw: Wydawnictwo Ministerstwa Obrony
Narodowej.
Smithyman, K. 1989. Selected Poems. Auckland: Auckland University
Press. Online http://www.smithymanonline.auckland.ac.nz/
Wedde, I, & H. McQueen. 1985. The Penguin Book of New Zealand Verse.
Auckland: Penguin Books.
CHAPTER THREE
THE POWER OF VISION

IN MYTHOLOGICAL THINKING:
‘TO SEE THE FORBIDDEN’ IN JAPANESE
MYTHS AND FAIRY TALES
MARINA SHCHEPETUNINA
Abstract. In a mythological worldview, the act of seeing is attributed to

power, and this power can be creative as well as destructive from our
human perspective. In myths we come across stories in which the act of
vision is closely connected to creation. Thus, in Japanese mythological
accounts of the Kojiki, when two deities descended to the first island
Onogoroshima in order to give birth to the land and myriad deities, the first
thing they did was “they saw to the erection of an heavenly august pillar,
they saw to the erection of an hall of eight fathoms” (Chamberlain 1982:
22). The phrase “saw to the erection” is Chamberlain’s translation of the
Japanese word mitatete, which can be interpreted as “to build by seeing”
(Ermakova 1991, Ermakova 1996). Another example of the creative power
of vision is that in many cultures the sun and moon are born from a deity’s
eyes, as in Japanese and Egyptian mythology, etc. On the other hand, there
are stories in which it is forbidden to look at something, and transgression
of the prohibition leads to undesired results. In this paper we examine the
meaning of vision in mythological thinking, with the major focus on
episodes where it is forbidden to look at something. Starting with the
ancient scientific understanding of sight, this paper offers insight into the
power which myths attributed to a look, and to the reasoning behind the
look taboo. The subject of this study is mainly ancient Japanese mythology,
as it is represented in the mythico-historical writings of the Kojiki ‘Records
of Ancient Matters’ (712) and Nihon shoki ‘Chronicles of Japan’ (720) as
well as Japanese fairytales. We also enrich our material with some well-
known stories from Greek mythology and related examples from African
mythology and folklore (Berezkin 2013).
26 Chapter Three
Introduction
In general, in primitive societies, mythological thinking commonly regarded
the visual acquisition of reality as the primary mode of perception, which
presupposed a special role of the look and vision in the religious domain
and their particular usage and meaning in magic and foretelling (Ermakova
1991). “In ancient times the main supernatural power of vision was its
ability to influence the object with the purpose of its stabilization,
pacification or empowering” (Ermakova 1991). One of the first attempts to
explain this power, to understand what vision is and to give it a scientific
explanation, is claimed to be the extramission theory of vision (Vavilov
1941). It is recognized that ancient Greek philosophy is rooted in a
mythological worldview (Asmus 1965). Plato (483-348) and Pythagoras (c.
532 BC), who preceded him, were the first to seriously develop this theory.
For Plato, the substance emitted from the eye was a kind of gentle "visual
fire" that flowed out of the pupil, which combined with light emanating
from the sun created a "body of vision". These lights would touch objects
and thereby generate a medium between the object and the viewer that
allowed aspects of the object to contact the soul (Ashbaugh 1988). In
Ashbaugh (1988) we find Plato’s theory as follows:
And of the organs they [gods] first contrived the eyes to give light … So
much of fire as would not burn, but gave a gentle light, they formed into a
substance akin to the light of every-day life; and the pure fire which is
within us and related thereto they made to flow through the eyes … When
the light of day surrounds the stream of vision, then like falls upon like,
and they coalesce, and one body is formed by natural affinity in the line of
vision, wherever the light that falls from within meets with an external
object. (Ashbaugh 1988: 8)
In the way that light and vision create the image of an object, in myths
the act of vision can create the object itself. Let us trace this idea in
Japanese myths.
Creative power of vision

The first example of a deity creating something by seeing is in the very
beginning of the myth story. Two deities, Izagi-no-mikoto and Izanami-
no-mikoto, descended to the first island Onogoroshima in order to give
birth to the land and myriad deities, and before doing so “they saw to the
erection of an heavenly august pillar, they saw to the erection of an hall of
eight fathoms”. The phrase “saw to the erection” is Chamberlain’s translation
The Power of Vision in Mythological Thinking 27
of Japanese mitatete, which can be interpreted as “build by seeing”

(Ermakova 1991, 1996). The island Onogoroshima was created by these
two deities when they stood “on the Heavenly bridge”, put a spear into the
ocean, and the drops from the spear consolidated and formed the island.
And then the two deities descended to the island and “saw to the erection”
of the first objects—the pillar and the hall, which we interpret as they
created them by their power of vision. After that they conducted a
marriage ritual and gave birth to other islands and gods as a man and a
woman. This is the beginning of the Kojiki story, and here we can observe
three different motifs of creation: the first is of the land taken from the
ocean, which resembles the earth diver and floating earth motifs, spread in
Indonesia (Obayashi 1975), the second is creation through vision, and the
third—giving birth as a human, as outlined above.
Another instance of this kind of creative power, as claimed by
Ermakova (1991), is the so-called kuni-mi, “seeing the country” or
“looking at the country” rite. Typically this rite was performed by an
Emperor, who, standing on some high place—a hill or a mountain—
looked out over his territory. One of the meanings of this act was to search
for fertile land to be used for agriculture (Origuchi 1955). Ermakova
(1991) suggests that there was also another meaning: this act helped to
establish the peace and stability of the country. It should be noted that
most of the kunimi songs in Manyoshu and Kojiki praise fertile and
prosperous land. One of such songs, composed by Emperor O-jin, appears
in Kojiki, section CVI:
And as I look on the Moor of Kidzu in Chiba

both the hundred thousand-fold abundant
house-places are visible and the land’s
acme is visible (Chamberlain 1982: 297)
Emperor O-jin composes this song while observing the country from a
hill in Uji. It is also important to note that Emperors in Japan, according to
Kojiki and Nihonshoki, are descendants of the sun goddess Amaterasu.
The story conveys that the grandchild of Amaterasu, Ninigi-no-mikoto,
descends to the Plain of Reeds (the land of Yamato) to rule it, and the first
Emperor of the human era, Emperor Jimmu, is his descendant. The
Emperor looking down from a mountain is like a sun shedding light on the
land. According to Ermakova (1991), “looking to the country from the
mountain Ame-no-Kaguyama thus is equivalent to the light of the sun
from the Heaven Mountain, which is bringing the order and vital power to
the world.” Here we should remember that for ancient people light and
vision are of the same nature, as was mentioned above.
28 Chapter Three
The light emanating from an eye was considered to be of the same

nature as that of the sun or the moon. There is evidence suggesting that the
connection of the sun and the moon to the eyes in many mythological
systems is not coincidental. People connected eyes with light. The sun
being born from one eye and the moon from another is a widespread motif.
According to Witzel (2012), this motif is found in Chinese, Indian, and
Japanese mythologies, among others. Let us examine Japanese mythological
accounts. Both in Nihonshoki and in Kojiki, when Izanagi-no-mikoto
returns from the netherworld, he undertakes a purification ritual, and upon
washing his one eye the Sun goddess Amaterasu was born; upon washing
the other eye, the Moon god, Tsukiyomi, was born. Sun and moon are
associated with eyes and light. If we have a look to the corresponding
versions of the birth of the sun and the moon from other variants recorded
in Nihonshoki, we can see not only this connection, but a directly spoken
creative power of vision. In the following story, the deity Izanagi, with the
intention to create the ruler of the world, makes the Sun and the Moon
deities by looking into a mirror.
In one writing it is said: - “Izanagi no Mikoto said: “I wish to procreate

the precious child who is to rule the world.’. He therefore took in his left
hand a white-copper mirror, upon which a Deity was produced from it
called Oho-hiru-me no Mikoto. In his right hand he took a white-copper
mirror, and forthwith there was produced from it a God who was named
Tsuku-yomi-no Mikoto. Again, while turning his head and looking
askance, a God was produced who was named Sosa no wo no Mikoto.
Now Oho-hirume no Mikoto and Tsuku-yomi no Mikoto were both of a
bright and beautiful nature, and were therefore made to shine down upon
Heaven and Earth. But Sosa no wo’s character was to love destruction,
and he was accordingly sent down to rule the Nether Land. (Aston 1972:
20)
The reflection of Izanagi’s eyes became the sun and the moon. His
intention was to create a ruler of the world, and by his look he created two
deities of “a bright and beautiful nature, and on the other hand, a deity
who was then sent to the underworld. What kind of look produces the
former? One is described as “looking askance”. When he looks straight he
produces the rulers of Heaven, and “while turning his head and looking
askance” he produces a chthonic creature. Similarly, in many mythologies
and cultures it is forbidden to look back, for it may cause bad fortune.
(Ermakova 1991)
For us, from the perspective of ancient conceptions of vision, the
crucial point is that which emits from the eye can influence an object, and
that influence or power can be creative as well as destructive. In the case
of a destructive power, it does not mean that we can destruct something by

looking, but we rather stabilize the situation and “freeze” what we see,
make it real.
In the next paragraphs we will explore the negative power of the look
in the context where it is prohibited to look at something.
Prohibition as motif
The taboo motif is a basic motif of countless folktales occurring around
the world in which the life, happiness, success, or failure of the characters
depends upon the observation or violation of some taboo. The entirety of
chapter C of Stith Thompson’s Motif-Index of Folk Literature is devoted
to the taboo motif in close to a thousand variants. There are taboos
connected with supernatural beings which define the rules of how to act
towards the supernatural, sex taboos, eating and drinking taboos, looking
taboos, speaking taboos, touching taboos, and so on. It should be noted
that all of these taboos prohibit some action that can influence an object, a
person or a deity. By touching or saying something you influence the
object, by eating or drinking influence yourself. And what about looking?
If we think of the materiality of the extramission theory introduced by
Plato, then we can understand the reasoning which may lead to the concept
that we can influence the object of the look. As discussed in the previous
paragraph the look in mythological thinking was endowed with the power
to influence an object. Further, we are going to examine the stories with
the prohibition to look. It is a taboo, and when something is tabooed then
it means that the action has sacred, magical power. By examining such
stories we are going to see what that power is and what is subjected to
such a taboo and why.
The looking taboo motif appears in different stories worldwide.
“Looking at the forbidden object or person causes its loss.” In Stith
Thompson’s Motif-Index of Folk Literature we come across the
prohibition to look at something as motifs C310- C320. The story
develops by first establishing that it is forbidden to look at a protagonist
doing something, then the prohibition is broken, and tragedy follows.
Probably the mostly well-known example of this taboo in European
culture is the Greek myth of Perseus and Medusa. Medusa, one of three
Gorgon sisters, is a female monster with a hideous face and snakes instead
of hair, who would turn to stone anyone who looked directly at her. The
hero Perseus beheaded her while looking at her image in a mirror. Here,
Medusa is a chthonic creature, which it is forbidden to look at (Agbunov,
1993).
30 Chapter Three
In Japanese myth we come across the prohibition to look motif in two

stories: that of Izanagi and Izanami, which is followed by an explanation
of the origin of death, and later the story of Toyotama hime, who prohibits
looking at her while she is giving birth to a child. In Japanese myths this
prohibition is connected to death and birth, and it should be noted that
these two stories share similarities in their structure. The common line of
these stories is: (1) a female deity enters a hut; (2) she prohibits a male
deity from looking at her; (3) the male deity breaks this prohibition; (4) the
female deity cannot remain in this world and is banished forever to
another world; and (5) the border between the worlds is established or the
passage between the two worlds is closed. These two ontological stories
have not only the look taboo in common, but their structure as well. In the
following sections we are going to explore these contexts accordingly.
The origin of death

The first story where the looking taboo appears in Japanese mythological
accounts Kojiki and Nihonshoki is of the male deity Izamagi-no-mikoto
following his wife and younger sister Izanami-no-mikoto to the Land of
Hades to bring her back to this world. When he came there, “from the
palace she [Izanami-no-mikoto] raised the door and came out to meet
him.” Izanagi-no-mikoto asked her to come back saying: “Thine
Augustness my lovely younger sister! The lands that I and thou made are
not yet finished making; so come back!” She answers that, although she
has eaten the food of Hades she wishes to return and will discuss it with
the deities of Hades and tells him not to look at her. “Having thus spoken,
she went back inside the palace; and as she tarried there very long, he
could not wait. So … he lit one light and went in and looked. Maggots
were swarming, and [she was] rotting, and in her head dwelt the Great-
Thunder, in her breast dwelt the Fire-Thunder …; - altogether eight
Thunder-Deities had been born and dwelt there.” (Chamberlain 1982: 40-
42). Having seen this Izanagi-no-mikoto fled back, was chased by the
Ugly-Females-of-Hades, then by the Thunder Deities. “Last of all his
younger sister Her Augustness the Princess-Who-Invites [Izanami-no-
mikoto] came out herself in pursuit. So he drew a thousand-draught rock,
and [with it] blocked up the Even Pass of Hades, and placed the rock in
the middle; and they stood opposite to one another and exchanged leave-
takings; and Her Augustness the Female-Who-Invites said: “My lovely
elder brother, thine Augustness! If thou do like this, I will in one day
strangle to death a thousand of the folks of thy land.” Then His
Augustness the Male-Who-Invites replied: “My lovely younger sister,
Thine Augustness! If thou do this, I will in one day set up a thousand and
five hundred parturition-houses.” In this manner each day a thousand
people would surely die, and each day a thousand and five hundred people
would surely be born. ” (Chamberlain, 1982: 44-45)
This story is found in Nihonshoki in eight variations, with the common
structure of: (1) Izanagi-no-mikoto follows Izanami-no-mikoto to the Land
of Hades; (2) a female deity enters a “hut” or a “palace” to get ready to
come back; (3) she prohibits a male deity from looking at her; (4) the male
deity breaks this prohibition; (5) Izanagi-no-mikoto is chased to the border
between the worlds; (6) the female deity cannot remain in this world and is
banished forever to another world; (7) the origin of death for humans is
spoken about.
The looking taboo in these episodes is generally understood as a
death taboo, that it is forbidden to look at the filth of death (Kurano 1963;
Yamaguchi & Konoshi 1997). On the other hand, in the Land of Hades,
after her death Izanami-no-mikoto meets Izanagi-no-mikoto and in this
stage there is no prohibition to look at her. Izanami-no-mikoto imposes the
looking taboo when entering the “the palace” in order to get ready to come
back to this world, in other words when she is trying to come to life or
rejuvenate. When Izanagi-no-mikoto breaks this taboo and enters “the
palace” he sees that “maggots were swarming, and [she was] rotting”, i.e.
the filth of the death, and that “altogether eight Thunder-Deities had been
born and dwelt there”, in other words she gave birth to thunder deities. We
may say that the place Izanami-no-mikoto hides herself is, first of all the
place where a transition is undertaken, and the transition is what it is taboo
to look at. In this episode we can observe transition of three kinds – that of
death, rejuvenation and giving birth. All these processes share being a kind
of transition between worlds.
In “the palace” we see the death and birth motif, and in the end of the
episode this dichotomy is spoken about again. Izanami-no-mikoto and
Izanagino-mikoto establish the reasoning for humans to die and to be born.
The female deity, who is banished to stay in the netherworld “will in one
day strangle to death a thousand of the folks” and the male deity will “in
one day set up a thousand and five hundred parturition-houses,” which
means that one thousand five hundred people will be born. This story
establishes the ontological order of life and death, and shows that the dead
are to stay in the other world, people will die and be born. And the turning
point of the story was the broken looking taboo with the reason that
Izanagi no mikoto saw the otherness and the situation was stabilised in the
condition he saw it.
32 Chapter Three
The connection between the origin of death and the look, in the form
that people lose the ability for rejuvenation due to the violation of the
looking taboo, can be found in African, South-East Asian, and South
American mythologies (Beryozkin 2013). In his profound research Africa,
Migrations, Mythology. Areas of the spread of the folklore motifs in
historical perspective Berezkin (2013), points that in many African
mythologies the change of skin is a condition for the afterlife. And African,
Indonesian, Melanesian, and South American versions share a common
detail that people cannot shed skin anymore because a person’s relatives
did not recognise him/her in a new appearance or were bothered in the
moment of rejuvenation (motif H4A) (Beryozkin 2013: 33). There are
variants when a child sees an old man changing skin and tells about it but
for a prohibition from saying (Eve tribe, South Uganda), an old woman
was bothered while shedding skin by her grandson’s calling her (Lur tribe,
North Ugand) and others. It should be noted that in some examples this
bothering is looking. Let us bring up an example of such a story as it is
told in the Chagga tribe in Kenia and Tanzania.
Mother sent his child to bring water and started to shed the skin. The child
came before he was expected and saw his mother getting out of the old
skin. The mother died and lost their ability for rejuvenation. (Beryozkin
2013: 33)
The looking taboo, although not directly spoken, appears in the context
of rejuvenation. Other stories of the H4A motif tell us that the process of
rebirth is not to be disturbed and that if the person is not recognized in the
new appearance it leads to the loss of the ability to rejuvenate. We may
suppose that once the otherness, in the form of the shed skin or a new
appearance, is brought to reality it cannot be changed and the new order of
being, i.e. mortality, is established.
The Japanese myth of Izanagi-no-mikoto travelling to the other world
and African rejuvenation stories give different explanations to human
mortality, but they share the idea that the rejuvenation process is not to be
disturbed and that once seen the object keeps that form. The one should
not be seen in the process of transition or as belonging to another, chthonic
world. In Japanese mythology and fairy tales we encounter this idea in the
context of giving birth.
Giving birth and the looking taboo

In Japanese mythological accounts the looking taboo appears in two
stories – the first, as discussed above, is the myth about Izanagi-no-mikoto
following Izanami-no-mikoto to the Land of Hades, which tells of the

origin of death, and the second is about a female deity Toyotama-hime,
daughter of the sea god Watatsumi-no-kami, giving birth in a parturition-
hall. Toyotama-hime comes to the seashore to Hoori-no-mikoto from the
Sea Plane to give birth to his child. She explains the reason for doing so:
“me thought that the august child of an Heavenly Deity ought not to be
born in the Sea Plane”. Being a chthonic creature, she comes from the
other world and brings the child to this world. She starts building the
parturition-hall, can not finish it, enters it and imposes a prohibition,
saying: “Whenever a foreigner is about to be delivered, she takes the shape
of her native land to be delivered. So I now will take my native shape to be
delivered. Pray look not upon me!” (Chamberlain 1982: 152). Hoori-no-
mikoto violates this prohibition, by looking at her “at the very moment of
delivery, when she turned into a crocodile eight fathoms [long], and
crawled and writhed about” (Chamberlain 1982: 152). Hoori-no-miloto
was terrified and fled away. Toyotama-hime said that she wanted to come
here and back between the two worlds, but because Hoori-no-mikoto
looked at her, she was “shame-faced”, closed the “sea boundary” and went
to the sea realm.
Toyotama-hime prohibits a male deity from looking at her at the
moment of delivery, explaining that at that moment she would change her
form and it is not to be seen. The violation of the taboo leads to the female
deity staying in the form she was seen in, i.e. as belonging to the other
world and unable to travel between the worlds any more.
Japanese scholars offer several explanations for this taboo. Sakamoto
Taro (1967: 165-167) in his comments in the translation of the Nihon
shoki into modern Japanese introduces the totemism theory proposed by
Matsumoto Nobuhiro, who claimed that Toyotama-hime and Ninigino-
mikoto belonged to different totem groups. Toyotama-hime’s totem
animal was a crocodile or a dragon, and at delivery there was a necessity
to conduct a special ritual concerning this totem, and it was tabooed for
foreigners to look at it. Tanigawa (1981: 12), emphasising the connection
of Toyotama-hime with the sea, interprets this myth as a variation of a
worldwide story of a supernatural woman who is turned into an animal
while bathing in water and, after having been seen is banished to leave.
Yoshino (1990) claims that in ancient Japan there was a belief in a snake
ancestor and that a woman in delivery would turn to its prototype – the
snake and the baby would be born in the form of a snake too and all that it
was taboo from looking at.
Tanigawa (1981) and Yoshino (1990) explain why Toyotama hime has
to change her appearance but not the reason for the implication of the
34 Chapter Three
taboo itself, just stating that the scene was about being prohibited from
looking. We suggest interpreting this episode from the meaning of vision
perspective, then, in the common context where we think that we can
explain this prohibition through the idea that ancient people attributed
magic power to vision, that by looking at something one can influence it.
The structure of this story in the part of imposing taboo shares the
same elements with the story about Izanami-no-mikoto, which are: (1) a
female deity enters a hut; (2) she prohibits a male deity from looking at
her; (3) the male deity breaks this prohibition; (4) the female deity cannot
remain in this world and is banished forever to another world; and (5) the
border between the worlds is established or the passage between the two
worlds is closed. In both stories it is forbidden to look at a female deity,
and both at the moment of taboo belong to the Netherworld. Izanami
belongs to the land of Hades and Toyotamabime – to the Sea-Plane,
besides she refers to herself as a “foreigner”. In Nihonshoki she turns into
a dragon (Nihonshoki, main story) or a sea monster (Nihonshoki variants).
The common structure is also found in Japanese fairy tales about a
supernatural wife, namely a snake-wife. In the collection of Japanese fairy
tales Seki (1972) points out that this Snake-wife fairy tale is spread all
over Japan from Aomori prefecture to Hiroshima prefecture and is found
in 47 variations. He tells the generalised story as follows:
A snake was saved and then came to the house of its savor in the shape
of a beautiful young woman and became his wife. Then, when it is time to
bear a child she prohibits others from looking at her and hides in a
parturition hall or in a room. A man breaks this prohibition and she is seen
as a snake. As a result she has to leave the family, and also leaves behind
her eye. Then the eye was stolen, and one day the man meets a woman
who had only one eye. She gives him the second eye to look for the child.
All the stories which with the looking taboo in the giving birth context
tell about the prohibition from looking at a female at the moment of
delivery, emphasize her chthonic nature and result in the violation of this
taboo so that she is banished to stay in the form in which she was seen.
Conclusions
In the Japanese mythological accounts Kojiki and Nihonshoki we could
observe vision as a means of creation and, in the form of a looking taboo,
as a means of destruction of the current order of being.
Vision in mythological thinking was attributed with power to influence
and even to create, and thus deities could produce objects just by the act of
looking. Izanami-no-mikoto and Izanagi-no-mikoto “saw to the erection of
an heavenly august pillar” and a hall, Izanagi-no-mikoto created deities of

the sun and moon by looking at a mirror, and we can suppose that he
actually saw his own eyes, by looking back at the mirror he produced the
chthonic deity Susano-no-mikoto. The creation of sun and moon deities is
in the frame of Frazer’s theory of a magic similarity principle. The deity’s
eyes are compared to the sun and moon, and seeing his own eyes Izanagi-
no-mikoto brings to reality what he saw, and stabilizes the seen.
In the context of a looking taboo – in the Land of Hades as well as in
giving birth and rejuvenation stories the actor, who looks at the forbidden
does not undergo changes, but the subject of the look, the one who is seen
at the moment of transformation, is influenced and is not able to conduct
this transformation back and stays in the condition it was seen. We may
say that the actor is creating a new order of being by looking, or at least
stabilizes the object. Such stories suggest an archaic understanding of
vision as a powerful act to influence the subject it is directed to, rather
than an act of cognition, for cognition is directed to the actor.
The concept of vision was one of the major ideas of the pre-Buddhist
archaic worldview in Japan and we may suppose that in mythological
thinking the primary meaning of vision was that of creation and stabilizing.
References
Asmus, V. F. 1965. Istoroya antichnoi filosofii [The history of ancient
philosophy]. Moscow: Visshaya shkola.
Aston, W.G. 1972 [1924]. Nihongi : chronicles of Japan from the earliest
times to A.D. 697 (2 vols). Tokyo: Tuttle Publishing.
Agbunov, M. (Ed.) 1993. Antichnye mify i legendy: mifologicheskiƱ slovarү
[Ancient myths and legends: mythological dictionary]. Moskow: MIKIS.
Ashbaugh, A. F. 1988. Plato's theory of explanation: a study of the
cosmological account in the Timaeus. Albany: State University of New
York Press.
Berezkin, Y. E. 2013. Africa, migratsii, mifologiya. Areali rasprostraneniya
folklornih motivov v istoricheskoi perspektive. [Africa, Migrations,
Mythology. Areas of the spread of the folklore motifs in historical
perspective.] Saint-Petersburg: Nauka.
Chamberlain, B. H. 1982 [1882]. The Kojiki: records of ancient matters
(Trans.). Tokyo: Tuttle Publishing.
Ermakova, L. M. 1991. Vzglyad i zrenie v drevneyaponskoi slovesnosti.
[Look and vision in ancient Japanese literature], Sad odnogo tsvetka:
sbornik statei i esse. [A garden of one flower: a collection of articles
and essays]. Moskow: Nauka. Online:
36 Chapter Three
http://dironweb.com/klinamen/read10.html
Ermakova, L. M. 1996. Naming and Seeing Things in Early Japanese
Poetry. Proceedings of the 38th Permanent Altaistic Conference (PIAC)
Wiesbaden, 135-141.
Frazer, J. G. 1980 [1913]. The golden bough: a study in magic and
religion (vol.2), Taboo and the perils of the soul. London: Macmillan
Kurano, K. 1963. Kojiki. Tokyo: Iwanami Shoten.
ƿbayashi, T. 1975. Nihon shinwa no kǀzǀ [The Structure of Japanese
Myth]. Tokyo: Kǀbundǀ.
Origuchi, N. 1955. Origuchi Nobuo zenshu: Dai 14kan, Kokubungaku
[The complete works of Origuchi Nobuo: vol.14, Japanese literature].
Tokyo: Chuokoronsha.
Sakamoto, T. M. Inoue, S. Ienaga & S. Ono 1993. Nihon shoki. Tokyo:
Iwanami Shoten.
Seki, K. 1972 [1953]. Nihon mukashibanashi shusei: dai 2 bu kaku
mukashibanashi [Collection of Japanese fairy tales: part 2 -1, full-
scale fairy tales]. Tokyo: Kadokawa shoten.
Tanigawa, K. 1981. Ubuya ko [Thinking about the parturition hut]. In K.
Tanigawa & Y. Nishiyama (Eds.), Ubuya no minzoku – Wakasa wan ni
okeru ubuya no kikigaki [Traditions concerning parturition hut: notes
about parturition hut in Wakasa Bay], (pp. 1-27). Tokyo: Kokusho
Kankokai.
Thompson, S. 1955. Motif-index of folk-literature: a classification of
narrative elements in folktales, ballads, myths, fables, mediaeval
romances, exempla, fabliaux, jestbooks, and local legends, Revised
and enlarged edition. Bloomington: Indiana University Press.
Vavilov, S. I. 1941. Glaz i solntse: o svete, solntse i zrenii. [The eye and
the sun: about light, sun and sight]. Moskwa/Leningrad: Izdatelstvo
akademii nauk SSSR.
Witzel, M. 2012 The origins of the world's mythologies. New York:
Oxford University Press.
Yamaguchi Y. & T. Konoshi (Trans. into modern Japanese, comments)
1997 Kojiki, Nihon koten bungaku zenshu [anthology of traditional
Japanese literature], Tokyo: Shogakkan.
Yoshino, H. 1990. Yo, gogyou to ubuya no minzoku. [Principles of Yang
& the five elements and traditions of the parturition hut]. In M. Senda
(Ed.), Kanshinakai bunka to kodai nihon – dokyo to sono shuhen [East
China sea culture and ancient Japan: Taoism and its peripheral], (pp.
275-302). Kyoto: Jimbunshoin.
PART TWO:
COGNITIVE-SEMIOTIC STUDIES
CHAPTER FOUR
GRAMMAR IN THE MIND

IN RELATION TO VISION:
AN ANALYSIS OF NULL-SUBJECT LANGUAGES
JUNICHI TOYOTA
Abstract. This paper examines how vision and grammar can be correlated
to each other. Vision here is broadly interpreted, including mental images.
In other words, vision is not simply a visual perception, but our
visualisation of events when we read or hear is also included. A particular
linguistic phenomenon examined here is the so-called null-subject
language. A number of languages do not overtly express the grammatical
subject, making a sharp contrast against languages such as English which
always require the presence of a subject. It is argued here that visualisation
of events in the mind helps interlocutors to supplement a referent for the
grammatical subject in null-subject languages. A speaker knows what
he/she refers to in speaking or writing, but hearers may not be able to trace
a referent correctly. It has been commonly considered that this
phenomenon is based on discourse factors, such as information structure.
However, since we constantly visualise events in our minds, this vision in
the mind, or rather, the ability to visualise internally, is an important
feature in successful communication when using these types of language.
The null-subject language is, therefore, suited for speakers to express
themselves. This point is discussed at length in terms of linguistic
orientation (Durst-Andersen 2011).
Introduction
We see events and describe them in our verbal communication, and our
interlocutors also visualise these scenes in their minds. What seems to be
the easiest aspect to this type of description is that people describe
everything in detail, but due to the economy of communication, some
features can be omitted. Omissions of this kind may appear to be a
hindrance, but speakers nevertheless are able to hold a normal
40 Chapter Four
communication. In this paper, a group of languages with a subject omitted,

known as the null subject languages, is examined. We shall focus on the
visualisation of events to see how a normal communication is made
possible without an overtly mentioned subject, and how this is achieved in
a null subject language. Due to some inherent grammatical features, some
languages are better suited to expressing certain features in our
communication, such as a description of events or the expression of
speakers’ opinions, known as linguistic orientation. Vision is examined in
relation to these inherent grammatical features here.
This paper is organised as follows: various characteristics of the null
subject language are presented first, including its geographical distribution.
Then visualisation in the mind is discussed, involving cultural differences
that can form a cognitive frame. Following these, a new type of
classification of language, i.e. linguistic orientation, is presented, and it
will serve as a base for the main discussion in this paper. Linguistic
orientation is also discussed in relation to the expression of the subject.
Finally, other semiotic means are also considered in relation to the null
subject language and linguistic orientation. There seem to be some
regional patterns, and these patterns are discussed in terms of cultural
practice or cognitive patterns, such as the sense of emptiness.
Null subject languages

In communicating events or situations, we normally deliver or exchange
information about an actor (i.e. a doer of actions) and an undergoer (i.e. a
person/an object affected by an action). It has been stated (cf. Givón 1979:
152) that human beings are ego/anthropocentric, and we tend to view
events in relation to human beings, i.e. how we perform an action, how we
are affected by an event, etc. Thus, the actor tends to be a human being,
and if this is violated, a structure is overtly marked, such as the passive
voice, where the subject tends to be predominantly inanimate (Toyota
2008: 115-118). The term actor is an inclusive term, but in the
grammatical tradition the actor is often termed as subject, and we shall
follow this common traditional term in this paper.
Having stated the ego/anthropocentric nature of language, it may be
natural to consider that our languages overly state who did an action, i.e.
the subject. However, this is not always the case. Among grammatical
diversity, there is a group of languages often called the null-subject
languages. A decisive characteristic of these languages is, as its
terminology suggests, that the subject is omissible. An example from
Japanese can illustrate this type, as demonstrated in (1). In (1a), the subject
Grammar in the Mind in Relation to Vision 41
is omitted, and it can refer to any pronouns, as the translation suggests, but
it can be overtly stated, as in (1b). In addition, the expression of the
subject can change when it is expressed with a pronoun, its word order
may change, or the pronoun becomes a verbal affix. For instance, in (2),
the pronominal subject is used as a verbal affix. These types of language
form a sharp structural contrast with those languages that obligatorily
require the presence of the subject, whether it is a noun, a pronoun or a so-
called dummy subject (i.e. form words such as it or there in English,
without any referential contents).
Japanese
(1) a. Soto-ni de-ta
outside-to go-PST
‘(I/you/he/she/it/we/they) went out.’
b. Watashi-wa soto-ni de-ta
I-TOP outside-to go-PST
‘I went out.’
Hakha Lai (Tibeto-Burman, Myanmar, Dryer 2013b)

(2) a-kal-tsaƾ
3SG-go-PRF
‘He has gone.’
Table 1. Typological patterns of pronominal subject (Source: Dryer 2013b)
Patterns of expressing pronominal subject Number of sample

languages
Pronominal subjects are expressed by pronouns in subject 82 (11.5%)
position that are normally if not obligatorily present
Pronominal subjects are expressed by affixes on verbs 437 (61.5%)
Pronominal subjects are expressed by clitics with variable host 32 (4.5%)
Pronominal subjects are expressed by subject pronouns that 67 (9.4%)
occur in a different syntactic position from full noun phrase
subjects
Pronominal subjects are expressed only by pronouns in subject 61 (8.6%)
position, but these pronouns are often left out
More than one of the above types with none dominant 32 (4.5%)
Total 711 (100%)
Dryer (2013) analyses a sample of 711 languages in the world and

classifies how the pronominal subject is expressed. The distribution is
shown in Table 1. The most common structure has the pronoun as an affix
on verbs, as in (2) from Hakha Lai. This type makes up ca. 61% of all the
data. The null-subject languages make up only 8.6% of the data, and this is
42 Chapter Four
not a common grammatical pattern cross-linguistically, but it forms an

areal feature, occurring mainly in East and Southeast Asia and Australia,
as shown in Figure 1. In previous understanding, the nature of null-subject
languages was ascribed to a discourse factor, i.e. a subject referent is
retrievable contextually and there is no need to express it overtly. However,
judging from the typological distribution, whether the subject is overtly
expressed or not may not be so simply a matter of discourse, as discussed
at length in the remainder of this paper.
Figure 1. Distribution of optional pronouns in subject position (Source: Dryer

2013b)
Figure 2. Distribution of obligatory pronouns in subject position (Source:

Dryer 2013b)
In addition, it may be worth mentioning that there are other

terminologies like pro-drop in some theoretical approaches, such as
Chomskyan syntax. It has been claimed that there is an underlying
pronominal subject in pro-drop languages. However, note that the majority

of the languages use affixes, which are not deletable, to refer to the subject,
that the number of languages with an obligatory subject is not so high (cf.
Table 1) and that they tend to form an areal feature, as demonstrated in
Figure 2. The areas with a high concentration of these languages are
northern Europe and West Africa (i.e. the Mande languages). Previous
research does not mention the Mande languages much, but focuses on
languages in northern Europe, particularly English. This Anglo-centrism is
often observable in earlier research, and the notion of an underlying
subject is no exception. With all due consideration of typological data, it is
apparent that this notion can be misleading.
Visibility in the mind

Visualisation may be crucial in some areas of our cultures. Religion can be
one of them. Visualisation can help the spread of belief and it takes
advantage of vision to varying degrees, some using paintings and icons as
much as possible, and others banning visualisations of gods. Religious
images of the same god may vary from culture to culture, although the
followers’ devotion may be equally strong regardless of their cultural
background. For instance, Christianity, covering different parts of the
world, often depicts images of its god. The realisation of such an image is
heavily influenced by each culture. Thus, a common image of Jesus Christ
in Europe is of a Caucasian, but the ones found in Japan, on the contrary,
resemble the features of Buddha statues. A BBC programme Son of God,
aired in 2001, questioned this, and based on archaeological findings the
programme reconstructed what Jesus might have looked like ca. 2,000
years ago. The reconstruction reveals a man from the Middle East, which
causes quite a stir since the image does not depict a Caucasian, as often
observable in churches and biblical paintings in Europe, and a number of
people could not accept the image shown in the programme. This suggests
that people in one culture share a similar image of god.
Similar cases can be found in our verbal communication, whether it is
spoken or written. We normally visualise events or scenes described in
utterances. In other words, we ‘see’ events through communication. This
means that by verbally communicating we are evoking an image in each
other’s minds. Whether we can have an identical image is questionable,
and this is what has been discussed in psychology and cognitive linguistics
as a prototype or cognitive frame, i.e. the difference is influenced by each
individual’s personal experience in life (cf. Enfield 2002; Taylor 2003;
Croft and Cruse 2004, among others). Thus, for instance, consider Figure
44 Chapter Four
3. Those who are familiar with European culture normally assume that the
people in the picture are waiting for a bus at a bus stop. However, this may
not be the case in some other cultures: there may be no bus stop (i.e. a pole
or a shelter marking a place where a bus stops when it arrives) in some
cultures and buses may be caught on the street by waving at a driver. Thus
Enfield (2002: 234-236) describes the case of speakers of Lao (Tai-Kadai,
Laos), who normally consider the action of the people in Figure 3 as
‘standing close to a signpost.’ This is natural since they have no clear
image of a bus stop in their minds. Thus, in such cases, images evoked in
the mind may not be identical, let alone similar, which may lead to
misunderstanding.
Figure 3. People waiting at a bus stop
Mendelsun (2014) questions how we really ‘see’ everything in

literature, and lists several cases where we can unknowingly fool ourselves
into believing that we visualise and understand who characters in a novel
are, including their appearance. For instance, in Leo Tolstoy’s Anna
Karenina, do we really know what the main protagonist, Anna Karenina,
looks like? Mendelsun (ibid.: 19) states the following:
Most authors (wittingly, unwittingly) provide their fictional characters with

more behavioral than physical description. Even if an author excels at
physical description, we are left with shambling concoctions of stray body
parts and random detail (authors can’t tell us everything). We fill in gaps.
We shade them in. We gloss over them. We elide. Anna: her hair, her
weight - these are only facets, and do not make up a true image of a person.
They make up a body type, a hair color … What does Anna look like? We
don’t know - our mental sketches of characters are worse than police
composites.
Thus, due to the paucity of description of Anna’s appearance in the novel,

we construct an image by filling the gap left by fragments of information
scattered in the novel, such as the aristocracy, 19th century Russia, the
feudal system, etc. These pieces of information may suffice to identify
Anna’s clothing or her living environment, but her face is hard to construct,
and this is where visualisation is heavily influenced by our cognitive frame.
Concerning our cognition, vision and mental images can be considered
closely related, and since this has a certain impact on our daily life, it is
possible that languages are also affected to some extent. So, let us now
consider the null-subject language. It is possible that a whole scene of
events can be visualised, including a doer of events. This suggests that a
subject referent may be present in our mental vision and it happens to be
omitted in a verbal utterance. Nevertheless, when we turn our attention to
the functional aspect of grammar, the visualisation of scenes can prove to
be an important feature in understanding grammar, although this has been
overlooked in previous research.
Orientation and visibility of grammar

Classification of languages has been attempted based on different criteria,
such as tense-based and aspect-based languages (Bhat 1999), and the null-
subject language can be considered as yet another criterion. In his study,
Durst-Andersen (2011) proposes yet another classification, stemming from
a cognitive semiotic approach to language. He claims that we may be able
to classify languages into three groups, e.g. reality, speaker and hearer, and
terms this type of classification linguistic orientation. According to his
approach, each language has arranged its grammar to fit one of the three
orientations above. Thus, by looking at some inherent grammatical
structures, one may be able to tell to which orientation type a language
belongs. Each orientation has representative languages, e.g. reality
(Russian, Chinese, etc.), speaker (Bulgarian, Turkish, etc.) and hearer
(English, Swedish, etc.). Each type is made, i.e. oriented, for a specific
target, either reality/situation, speaker or hearer. The presence of certain
features can set each language into a certain orientation type.
Thus, in reality orientation, a speaker functions as a reporter present at
a situation. Thus, how a situation is described is important in this type and
grammar is equipped with various tactics to deal with this demand. For
instance, it is important to differentiate whether an event is still ongoing or
finished (i.e. aspectual distinction between perfective and imperfective).
We shall see more examples shortly in this section.
46 Chapter Four
A speaker in speaker-orientation functions as a commentator,

expressing his/her own view on events. Thus, what he/she experiences is
reflected in utterances, forcing grammar to be equipped with devices to
describe details of experience, such as evidentiality (cf. Aikhenvald 2004),
to differentiate whether he/she has a direct experience or not, and if he/she
does, how such experience was gained, e.g. though vision, sound, smell,
etc. An extreme case is found in a Papuan language, Fasu, where the
English phrase It’s coming has six distinct translations according to the
degree of evidentiality, as demonstrated in (3). Note that the ability to see,
i.e. vision, plays a role in distinguishing evidentiality in this language, e.g.
(3a). Thus, this type of orientation is concerned with modality, in
particular, distinctions in the indicative mood.
Fasu (Papuan, Foley 1980)

(3) a. aspere. ‘I see it [it’s coming].’
b. perarakae. ‘I hear it [it’s coming].’
c. pesareapo. ‘I infer it from other evidence [that it’s coming].’
d. pesapakae. ‘Somebody said so [i.e. it’s coming], but I don’t
know who.’
e. pesaripo. ‘Somebody said so [i.e. it’s coming], and I know
who.’
f. pesapi. ‘I suppose so [i.e. it’s probably coming].’
The third type, hearer-orientation, has an elaborate system of

identifying different types of information, such as new and old, referable
and non-referable, etc. This is encoded in the simple past tense (as
opposed to the perfective aspect) or articles (e.g. definite vs. indefinite).
These aid the hearer to decode details of information and identify whether
a referent is familiar to him or not. In this type, interlocutors consider
information as the basic unit. The speaker is second person-oriented, acts
as an informer and speaks with an intersubjective voice. In order to make a
reference, this type makes use of both indefinite and definite articles. And
the use and importance of the definite article is also shown in its historical
development. The common source for the definite article is demonstrative
pronouns (Heine and Kuteva 2002: 109-111), e.g. English the is derived
from the Old English demonstrative se (s.v. OED the dem. a. (def. article)
and pron.). Note, however, that speaker-oriented languages sometimes use
definite articles too, but with different purposes. Thus, for instance,
Bulgarian uses a demonstrative as a definite article, although there is no
indefinite article. Consider the examples in (4). In (4b), -ta is a clitic form
of the demonstrative, functioning as a reminder of referents to speakers,

not to hearers.
Bulgarian
(4) a. kniga ‘a book’
b. knigata ‘the book’
The characteristics described so far can be summarised in Table 2. The

representative languages listed here are taken from the Eurasia region, and
further research is definitely required to include languages from other
regions, such as Africa, to make the list more typologically viable.
Table 2. Characteristics associated with orientations

Reality-orientation Speaker-orientation Hearer-orientation
Representatives Russian, Chinese, Bulgarian, Turkish, English, Swedish,
etc. Japanese, etc. Danish, etc.
Basic unit Situation Experience Information
Speaker orientation Third person First person Second person
Speaker function Reporter Commentator Informer
Identification mark Aspect prominence Mood prominence Tense prominence
Expression of subject and orientation

Having seen a rough outline, let us take a closer look at how orientation
works in relation to vision, taking Russian as an example. Reality
orientation mainly deals with ‘here and now’ in conversation, which
presumes that interlocutors are familiar with their visible surroundings.
Thus, the grammar in this type distinguishes a physically visible situation
from a non-visible one. The Russian examples in (5) illustrate the point in
question. Both examples in (5) refer to possession, but (5a) with imeti
‘have’ is restricted to a case where abstract nouns are involved (i.e. non-
visibility), whereas (5b) deals with possession with concrete nouns (i.e.
visibility). Thus, possession is divided into two types according to whether
a possessed object is physically visible to the naked eye or not. Possession
in world languages can be classified into eight types (Heine 1993), and
note that some languages use more than one type. A clear distinction in the
construction of possession as seen in (5) indicates that visibility plays a
role in grammatical organisation in reality orientation.
Another feature illustrating a relationship between visibility and
grammar is negation. When a clause is negated, a case marking can change
from its affirmative counterpart, and this is most vividly shown in the
existential clause, as demonstrated in (6). Slavic languages, as well as
48 Chapter Four
older Indo-European languages, optionally use the genitive case for the
grammatical object when a clause with an accusative object is negated.
Consider the examples in (7). A traditional grammar requires a change in
the case marking, and the genitive form is an ideal choice (i.e. (7c)), but
the accusative case can be retained in modern grammar, i.e. (7b). However,
the existential clause in Russian forces a change in the case marking in the
negative clause, e.g. the affirmative clause in (6a) uses the nominative case
for an NP, but once it is negated, the genitive case has to be used, as in
(6b).
Russian
(5) a. Ya imeju mnenie
I have.PRS opinion.ACC
‘I have an opinion.’ (abstract noun)
b. U menja jest’ kniga
with I.ACC.SG exist.PRS book.NOM
‘I have a book.’ (concrete noun)
Russian
(6) a. Byla kniga
was book.NOM
‘There was a book.’
b. Ne bylo knigi
NEG was book.GEN
‘There was not a book.’
Russian
(7) a. Ja chital etu knigu
I read.PST.PRT.M this.ACC book.ACC
‘I read this book.’
b. Ja ne chital etu knigu
I NEG read.PST.PRT.M this.ACC book.ACC
‘I did not read this book.’
c. Ja ne chital etoy knigi
I NEG read.PST.PRT.M this.GEN book.GEN
‘I did not read this book.’
Yet another example is taken from marked case markings on the

subject. Consider the examples in (8). Both examples refer to the same
situation, but the subject differs in its case marking depending on whether
a speaker can directly experience the event (i.e. being able to observe it
visibly) or not at the time of utterance. If he/she can (e.g. while it was
raining), the nominative case is used as in (8a), but if he/she cannot (e.g.
after the rain), the nominative case is replaced by the instrument case as in
(8b). This may appear to be a case of evidentiality (cf. (3) from a Papuan
language, Fasu), since both cases refer to a distinction between first-hand
and non-first-hand experience. However, in Russian a speaker is forced to
make a choice between these options according to a real-time situation in
speaker’s environment. In other words, the direct experience in relation to
vision is crucial here. Timeliness is not always required in evidentility, and
this is what makes the Russian grammar a stereotypical case of reality
orientation.
Russian
(8) a. Dozhd’ smy-l pyl-’
rain-NOM wash.down-PST.PRT.M dust-ACC
‘The rain washed down the dust.’
b. Dozhd-yom smy-lo phy-’
rain-INST wash.down-PST.PRT.N dust-ACC
‘The dust was washed down by the rain.’ (lit. washed down
the dust by the rain.’)
Russian may be an extreme case of reality-orientation, but similar

grammatical behaviours based on visibility cannot be found in other
orientations. In addition, reality orientation does not distinguish between a
spoken and written register, since a spoken one is inherently concerned
with ‘here and now’, whereas a written register can transfer messages
beyond time and space (cf. Toyota 2009). In another type, speaker
orientation, a speaker visualises a situation in his/her mind, and the
grammar does not have to describe overtly details of the situation to others,
i.e. the grammar describes what is thought in a speaker’s head. These
languages may use a definite article or a demonstrative as a definite article,
but the indefinite article is not present, i.e. the definiteness is used in order
to remind a speaker, but not a hearer, of a referent, as already
demonstrated in (4) from Bulgarian. Furthermore, hearer-orientation has to
describe events in detail so that hearers can visualise and cognitively
process a situation expressed in utterances easily. Both definite and
indefinite articles are used in order for hearers to decode contextual clues
to sort out information.
English is an extreme case of hearer-orientation, in sharp contrast to
Russian. The English grammar is full of devices to evoke vision in a
hearer’s mind, which is not observable in other types of orientation. This
50 Chapter Four
may explain the grammatical peculiarities of some languages such as

English (Toyota 2012), and in this sense, the history of English can be
summed up as a history of making devices to conjure up a vision as close
as possible to the speaker’s in the hearer’s mind, i.e. a shift from a
speaker- to hearer-orientation.
Correlation with other semiotic resources

Earlier in Section 2, we saw that there is an areal feature related to the
null-subject language, and this type of grammatical feature is very
common in East and Southeast Asia and Australia. Similar distributional
patterns involving other grammatical features can be found. Toyota (2013)
shows a case of the absence of the overtly-marked future tense in the area
where the null-subject languages are commonly found. Consider the
distribution of the future tense in Figure 4. Note that the future tense
referred to here is the inflectionally marked one (i.e. the inflectional future
tense), as shown in (9) from Serbian, and example (9b) also shows a case
of the inflectional future tense without an overtly-expressed subject.
Serbian
(9) a. Ja þekam ovde
I.NOM wait.PRS here
‘I wait here.’
b. ýeka-üu ovde
wait-FUT.1SG here
‘I will wait here.’
Notes: white spot = no future tense; darker spot = inflectionally-marked future tense
Figure 4. Distribution of inflectional marking of the future tense (Source: Dahl
and Velupillai 2013)
Once the maps in Figure 1 (the null-subject languages), Figure 2

(languages with an obligatory subject) and Figure 4 (the future tense) are
combined, another distributional map is obtained, as shown in Figure 5.
The number of languages drops drastically, due to the fact that some
languages do not employ all the grammatical features, and those which do
use all the features under examination are represented in Figure 5. This
clearly shows an areal feature around East and Southeast Asia for the lack
of the inflectional future tense and the null-subject language.
Notes: black circle = inflectional future & obligatory subject; black square = no
inflectional future & subject optional; dark grey circle = no inflectional future &
obligatory subject; pale grey circle = inflectional future & subject optional
Figure 5. Distribution of future tense and obligatory subject combined (Source:
Dryer and Haspelmath 2013)
The correlation does not stop here, and there are other features that
show a similar areal distribution. Concerning counting, for instance,
languages can be divided into classifier and non-classifier languages. The
difference between them is that the non-classifier languages distinguish
mass nouns from countable nouns and use a classifier for countable nouns.
Classifier languages, on the other hand, treat mass and countable nouns in
the same way, using classifiers for both types of nouns, as summarised in
Lyons (1977: 463):
[Non-classifier] languages which grammaticalize the distinction

between entity-denoting nouns and mass-denoting nouns tend to
draw a sharp syntactic distinction between phrases like “three
men” on the one hand, and “three glasses of whisky,” on the other.
Classifier languages do not: they treat enumerable entities and
enumerable quanta in much the same way.
52 Chapter Four
Toyota and Kovaþeviü (2013) identify a regional distribution of these

counting types, and claim that classifier languages are concentrated in East
and Southeast Asia, as shown in Figure 6. They argue that this tendency
can be ascribed to the cognitive and cultural difference discussed by
Nisbett (2003) and Nisbett and Masuda (2007), i.e. Europeans see the
world analytically, while Asians tend to see it holistically. Nisbett and his
colleagues analyse picture recognition using eye-tracking, and have
identified a clear behavioural pattern according to participants’ cultural
backgrounds.
Notes: dark circle = numeral classifiers absent; grey circle = numeral classifiers
optional; white circle = numeral classifier absent
Figure 6. Distribution of numeral classifiers (Gil 2013)
How can we explain the various characteristics and geographic

distribution of the null-subject languages and those that obligatorily
require the overt marking of the subject, and other grammatical
characteristics exhibiting similar geographic distributions?
Since speakers have a communal attitude towards life in East and
Southeast Asia, their worldview serves as a common cognitive frame. So
they assume that much is already known to their interlocutors, and
therefore that there is no need to mention or specify participants and
events. Thus, languages spoken in the area where the null-subject
languages are found commonly lack definite articles, too. Consider Figure
7 and compare it with Figure 1. Languages in Western Europe normally
possess the definite article, whereas languages spoken in East and
Southeast Asia generally lack the article, and if present, it is normally the
indefinite article. In this environment, speaker-orientation can flourish, i.e.
speakers safely assume that hearers have some background knowledge and
can leave successful communication to hearers’ interpretation when using
such knowledge. This is not observable amongst reality-orientation and

hearer-orientation: reality-orientation needs to demonstrate a real situation
without using resources taken from the hearers’ knowledge, and hearer-
orientation has to describe much in detail to hearers, without taking
advantage of any common knowledge.
Notes: black circle = definite article; white square = no definite, but indefinite
article; white circle = no definite and indefinite article
Figure 7. Distribution of definite and indefinite articles (Dryer 2013a)
It is worth mentioning that hearer orientation may correspond to Hall’s

(1976) low-context, providing a lot of contextual clues in the grammar.
Hall’s high-context, on the other hand, can be represented by speaker
orientation, expecting hearers to do much of the interpretation. His socio-
psychological approach has been studied in various disciplines, but this
difference can be indicated by grammatical characteristics discussed here
too.
Concerning vision, what is absent in grammar may not be absent
altogether. In speaker orientation, speakers expect hearers to fill gaps
created by linguistic description. However, speakers and hearers can
visualise a scene from a verbal communication, and what appears to be
missing in grammar, such as the subject, may well be present in the
interlocutors’ mind’s eye. Earlier it was stated that interlocutors share a
common frame, and this frame, concerning vision, includes pictorial
images. Considering this, linguistic orientation can be regarded as
different means for representing pictorial images in the mind, i.e. reality-
orientation distinguishes whether entities in a picture are visible or not,
speaker-orientation relies on a common mental frame in the hearer’s mind
to complete pictures, and hearer-orientation explains entities in a picture
verbally to hearers.
54 Chapter Four
Conclusions
This paper has examined the nature of the null subject language, and how
the omission of the subject is functionally viable. It has been argued that
due to the inherent nature of grammar, known as linguistic orientation,
some languages are focused on speakers’ expressibility (i.e. speaker-
orientation), resulting in the lack of overt marking of some grammatical
features, and in the case of the null subject language, this missing feature
is the subject. This may appear to be disadvantageous in communication,
but speakers of these languages can successfully communicate. What
allows the null subject language to be fully functional is, as argued here,
visualisation in the mind, i.e. interlocutors visualise scenes of events
internally which supply the missing subject referent in utterances.
Visibility or visualisation plays an important role in grammatical
organisation, and vision is not simply a cognitive input. It seems that
visualisation in a hearer’s mind (i.e. hearer orientation) is a distinct type of
grammatical organisation, making a sharp distinction against reality- as
well as speaker-orientation. This may explain extreme cases of inherent
grammatical structures and functions associated with them, such as
Russian (reality-orientation) and English (hearer-orientation). Speaker-
orientation is mainly concerned with the expressiblity of speakers, and
grammar is organised for this function. Hearers use internal visualisation
of events to cope with obvious omissions of some grammatical items.
Thus, vision-related issues can be incorporated into grammar to
understand better the functions of human languages.
References
Bhat, D. N. S. 1999. The Prominence of Tense, Aspect and Mood.
Amsterdam: John Benjamins.
Croft, W. & D. A. Cruse 2004. Cognitive linguistics. Cambridge:
Cambridge University Press.
Dahl, Östen & V. Velupillai 2013. The future tense. In M. S. Dryer & M.
Haspelmath (Eds.), The world atlas of language structures online.
Leipzig: Max Planck Institute for Evolutionary Anthropology.
(Available online at http://wals.info/chapter/67, Accessed on 2016-08-
10.)
Dryer, M. S. 2013a. Definite articles. In M. S. Dryer & M. Haspelmath
(Eds.), The world atlas of language structures online. Leipzig: Max
Planck Institute for Evolutionary Anthropology. (Available online at
http://wals.info/chapter/37, Accessed on 2016-08-26.)
Dryer, M. S. 2013b. Expression of Pronominal Subjects. In M. S. Dryer &

M. Haspelmath (Eds.), The world atlas of language structures online.
(Available online at http://wals.info/chapter/101, Accessed on 2016-
08-09.)
Dryer, M. S. & M. Haspelmath (Eds.) 2013. The world atlas of language
structures online. Leipzig: Max Planck Institute for Evolutionary
Anthropology.
Durst-Andersen, P. 2011. Linguistic Supertype. Berlin: Mouton de Gryuter.
Enfield, N. J. 2002. Cultural logic and syntactic productivity: associated
posture constructions in Lao. In N. J. Enfield (Ed.), Ethnosyntax:
Explorations in grammar & culture (pp. 231-258). Oxford: Oxford
University Press.
Gil, D. 2013. Numeral classifiers. In M. S. Dryer & M. Haspelmath (Eds.),
The world atlas of language structures online. Leipzig: Max Planck
Institute for Evolutionary Anthropology. (Available online at
http://wals.info/chapter/55, Accessed on 2016-08-31.)
Givón, T. 1979. On Understanding Grammar. New York: Academic Press.
Hall, E. T. 1976. Beyond Culture. Oxford: Anchor Books.
Heine, B. 1993. Possession. Cambridge: Cambridge University Press.
Heine, B. & T. Kuteva 2002. World Lexicon of Grammaticalization.
Cambridge: Cambridge University Press.
Mendelsund, P. 2014. What We See When We Read. New York: Vintage.
Nisbett, R. E. 2003. The Geography of Thought. New York: Free Press.
Nisbett, R. E, & T. Masuda 2007. Culture and point of view. Intellectica:
Revue de l’Assoiciation pour la Recherche Cognitive, 2-3, 153-172.
OED (Oxford English Dictionary). 1989. 2nd edn. Oxford: Clarendon
Press.
Taylor, J. R. 2003. Cognitive Grammar. Oxford: Oxford University Press.
Toyota, J. 2008. Diachronic Change in the English Passive. Basingstoke:
Palgrave.
—. 2009. ‘Orientation reflected in register: from historical perspectives.’
Discourse and Interaction, 6, 45-61.
—. 2012. ‘Dialect mixing as a language contact in the history of English.’
In R. Trušník et al. (Eds), Theories and Practice: Proceedings of the
third international conference on Anglophone Studies (pp. 95-110).
Zlín, Tomas Bata University.
CHAPTER FIVE
MOTIVATIONS FOR COUNTER SYMBOLISM:

A CASE OF DEMONSTRATIVES
IN CUSHITIC LANGUAGES
JUNICHI TOYOTA
Abstract. Sound symbolism represents an iconic or indexical relationship

between sounds uttered and objects referred to by these sounds. What is
involved in this relationship can vary, i.e. perhaps physical size is what is
most commonly associated with sound symbolism, but other categories
such as physical distance can be expressed, e.g. /i/ represents the smallest
distance (i.e. proximal) and /u/, the largest distance (i.e. distal). However,
this relationship does not hold in some languages. A notable exception for
this tendency is demonstratives in a Cushitic language, Somali. The vowels
have a reverse pattern from what sound symbolism commonly predicts, but
since this makes an opposing pair, it may be possible to identify certain
motivations. This paper proposes a possible reason for cases like the
Somali demonstrative, termed here as counter-symbolism, based on
speakers’ vision.
Introduction
Sound symbolism has attracted attention from different disciplines, and
thanks to previous research, a general pattern has been identified.
Concerning the theme of this paper, spatial distance can be represented by
sounds, particularly vowels, i.e. the proximity by high front vowels such
as /i/ and the distance by high back vowels such as /u/. In spite of this
crosslinguistic tendency, there are indeed some exceptions, although they
are not so common. A case examined in this paper, the demonstratives in
Somali, belongs to the exceptions, and this paper investigates how this
exception emerged in this language, and an attempt is made to identify
possible motivations for such an exceptional case. For this, vision seems to
58 Chapter Five
play a role, and how an object is perceived seems to influence the choice
of vowels.
This paper is organised as follows: demonstratives in the Cushitic
languages, in particular Somali, are first presented, and the data presented
here serve as a base for the main analysis. Following this, we examine the
demonstratives in terms of sound symbolism, and identify exceptional
cases. Finally, a possible motivation for having exceptional cases in
Somali is postulated.
Demonstratives in the Cushitic languages

The term demonstrative can entail several subtypes. Its basic function is to
distinguish a referent from others, and it is normally deictic referring to
proximity or distance from a speaker or interlocutors. What is referred to
is typically spatial, as in the English this book and that book, but it can be
temporal when used metaphorically, as in the English this morning and
that morning. Its deictic usage can also be extended to discourse, often
discussed under another term, anaphora. A common distinction of distance
is binary (here and there) or ternary (here, there and intermediate), but
other complex cases can be found (cf. Diessel 2013, as well as Table 1
from Somali and Table 2 from Iraqw, both Cushitic). Apart from these
basic features, there are, however, a number of cases which are language
specific, and vision can play an important role. For instance, Trio (Carib,
Surinam) has the usual set of demonstrative pronouns, but there is a
special one, denoting ‘not visible, but audible’, as demonstrated in (1),
where the demonstrative pronominal prefix më- serves this purpose.
Notice the use of the anaphoric inanimate demonstrative pronominal suffix
irë-.
Trio (Carib, Carling 2004: 418)

(1) irë-mao i-n-muku-ru-ja
DP.INAN.ANA-TEMP 3POSS-3O-bear.NOM-POSS-GOAL
t-ï-ponoh-po-e a-kï më-kï
COREF-TR-tell-CAUS-NF Q-ANIM DP.AUD.NVIS-ANIM
tïï-ka-e
COREF.1TR-say-NF
‘Then his son asked ‘who is that?’ he said.’
When it comes to the Cushitic languages, one can find a rather

complex system for demonstratives. A set of examples taken from Somali
is shown in Table 1. As this table demonstrates, the distinction in terms of
distance is quaternary, proximal, medial, distal and far distal. The deictic
Motivations for Counter Symbolism 59
centre here is the speaker. However, the case of Iraqw, as demonstrated in

Table 2, involves both a speaker and a hearer as the deictic centre. Thus, in
terms of distance, Iraqw has a ternary deictic system.
Table 1. Demonstratives in Somali (Saeed 1999: 113)

Proximal Medial Distal Far distal
(close to (further away (in the middle (in the far
speaker) from speaker) distance) distance)
M -kán -káa(s) -kéer -kóo
F -tán -táa(s) -téer -tóo
Table 2. Demonstratives in Iraqw (Mous 1993: 90-91)

‘near the speaker’ ‘near the addressee’ ‘near neither of them ‘far away’
but still visible’
-i/-ká -síng -qá -dá
Some Cushitic languages, Somali in particular, have developed a

system of anaphora from demonstratives, as shown in Table 3. The
distinction in this case is a simple binary one, and some scholars, such as
Saeed (1999: 112) consider it as a grammatical item comparable to the
articles found in some of the Indo-European languages such as English.
This is a common developmental path found elsewhere. See Heine and
Kuteva (2002: 109-112) for examples. Some Cushitic languages, such as
Rendille (Oomen 1978), Boni (Sasse 1981), Dirayta (Hayward & Saeed
1984), and Somali (Saeed 1999) obligatorily require a focus or topic
markers. This means that speakers of these languages must show
sensitivity to the information structure at all times. The examples in (2)
demonstrate a case in Somali. In these examples, the focus markers báa in
(2a) and wáxa in (2b) are used, and with báa, the focused element must
come before a VP, while with wáxa, after a VP. In both examples, what is
focused on is Soomaaliyád ‘Somali woman’. See Saeed (1999: 189-96) for
a detailed description of these markers. Under these circumstances, the
grammar of these languages shows sensitivity to the discourse structure by
possessing an intricate system of focus, and it is not a surprise that the
demonstratives have developed into anaphora in Somali. In this article, we
do not make a finer distinction, and treat items in Table 1 and Table 3
collectively as demonstratives in Somali.
Table 3. Articles/demonstratives in Somali (Saeed 1999: 112)

Non-remote Remote
M -ka -kii
F -ta -tii
Somali (Saeed 1999: 190)
60 Chapter Five
(2) a. Soomaaliyád bàad tahay

Somali.F bàa.you are
‘You are a Somali woman.’
b. Wáxaad tahay Soomaaliyád
wáxa.you are Somali.F
‘What you are is a Somali woman.’
Furthermore, the demonstrative in the Cushitic languages is known for

its extension into the temporal expression, known as the nominal tense.
See Muysken (2008) and Nordlinger and Sadler (2004) for examples from
various languages. The way the nominal tense in the Cushitic languages
work is that the addition of a demonstrative does not create a spatial sense,
but a temporal one. In other words, the proximity in this case corresponds
to the present tense and its various aspectual variations, such as present
perfective or progressive, and the distance refers to the remoteness of time,
particularly in the past. The examples from Iraqw in (3b) and from Somali
in (4) demonstrate the nominal tense. The distant demonstrative -dá on the
noun hasama ‘house’, not on bal ‘day’, in (3b) refers to the distant past.
This example forms a sharp contract with (3a): notice that there is no tense
marker in these examples on the copula verb, and the presence/absence of
the demonstrative solely triggers the difference in tense. This is sometimes
known as an independent nominal tense-aspect-mood. Mous (1993: 90-91)
suggests that the demonstratives in Iraqw can be a tense marker, e.g. -dá
for present and -qá for past. The nominal tense marker provides temporal
information local to the NP, and it can operate independently of the tense,
aspect and mood of a clause. These markers are fully productive and they
can appear with all members of the nominal word class.
Iraqw (Mous 1993: 90, 124)

(3) a. iraqw a doohlite
Iraqw COP farmers
‘Iraqw are farmers.’
b. hasama-dá bal-á a aseemi
house-DEM day-DEM COP flat.roof.house
‘Those houses of that period were flat-roofed houses.’
The nominal tense in Somali, on the other hand, is an auxiliary tense

marker. Both examples in (4) carry the tense marker on the verb, but the
addition of the demonstrative provides additional information concerning
the noun a demonstrative modifies. Thus, the non-remote demonstrative -
da on ardáy ‘students’ in (4a) refers to the present, although the whole
predicate is in the past tense. In (4b), the remote demonstrative -dii forces
the past tense on the noun phrase, but the tense of the predicate is
determined by the marker on the verb, i.e. non-past. This type is known as
an independent tense-aspect-mood, i.e. nominal with a tense-aspect-mood
marker which is interpreted with respect to the higher clause within which
it is embedded. Scholars seem to have different opinions over whether the
nominal tense exists in Somali. Lecarme discusses this in Somali
extensively (e.g. Lecarme 1999, 2008), using demonstratives as tense
markers. Saeed (1999: 111-114), on the other hand, treats the same
markers as definite articles, i.e. Table 3.
Somali (Lecarme 1999: 335)

(4) a. ardáy-da baan kasin
students-DET.F.NREM FOC.NEG understand.PST
su’áah-aadii
question-DET.F.POSS.2SG
‘The students (who are present/I am telling you about) did
not understand the question.’
b. ardáy-dii wáy joodaan
students-DET.F.REM FOC.3PL are.present.NPST
‘The students (I told you about) are present.’
The use of the demonstrative in Somali is diverse and its application to

the temporal domain and discourse is quite visible, as the examples so far
have demonstrated. There is, however, something odd in the demonstratives
in Somali. What is not so visible here is a sound.
Sound symbolism and exceptions

Covered by a diverse usage, it is often overlooked, but the formation of the
demonstrative is rather peculiar in the Cushitic languages, particularly in
Somali. In order to clarify this point, one needs to look into sound
symbolism, i.e. the relationship between the vowel and a shape or distance.
Sound symbolism represents an iconic or indexical relationship between
sounds uttered and objects referred to by these sounds. What is involved in
this relationship can vary, i.e. perhaps physical size is what is most
commonly associated with sound symbolism, but other categories such as
physical distance can be expressed. Traumüller (1994), for instance, finds
a certain tendency in the demonstratives (32 out of 37 sample languages)
between the quality of vowels and distance referred to, e.g. /i/ represents
the smallest distance (i.e. proximal) and /u/, the largest distance (i.e. distal).
62 Chapter Five
This type of relationship is often observable in the diminutive and

augmentative, as demonstrated in (5) and (6), respectively. In (5) from
Yagua (Peba-Yaguan, Peru), the addition of a diminutive marker -déé
makes the size of the referent smaller, and in (6) from Spanish, the suffix -
lon creates negative or undesirable connotations. Note that the diminutive
form often carries a sense of endearment. Concerning the formation of
these forms, Payne (1997: 110) notes that ‘[t]here is an apparently
universal iconic tendency in diminutives and augumentatives: diminutives
tend to contain high front vowels, whereas augumentatives tend to contain
high back vowels.’ A finer distinction of vowels is listed in Figure 1.
These examples follow this pattern well, and when the iconic relationship
is extended to space, the high front vowels represent closeness, and the
back high vowels, distance.
Yagua (Peba-Yaguan, Peru, Payne 1997: 110)

(5) a. quivą̗ą̗
‘fish’
b. quivąądéé
‘little fish’
Spanish (Payne 1997: 110)

(6) durmi-lon
sleep-AUG
‘sleepyhead/lazybones’
Figure 1. Likelihood of vowels representing a distance from a speaker (source:

Johansson 2011: 29)
Some previous works such as Traumüller (1994), Zlatev & Andrén

(2009) and Ahlner & Zlatev (2011) have proposed several instances of
iconicity or indexicality in counter-symbolism. For instance,
demonstratives in a Chadic language, Hausa, differentiate the distance in
terms of consonants, not of vowels (apart from a tonal difference). Hausa
has a ternary distinction of distance, and the ones near interlocutors have
the alveolar nasal /n/ as an initial consonant, while the ones denoting
distance have the velar plosive /k/ as an initial consonant. The amount of
closure in the vocal cavity in producing the nasal sound is much
minimised, but for the velar, it is relatively open, as demonstrated in
Figure 2, where the size of the circle represents the space in the cavity.
This varying degree of openness may correspond to a proximal (closure)
and distal (open) distinction (cf. Ikegami and Zlatev 2007). In addition to
this, the opening of the mouth can be an iconic sign, i.e. a small opening
for smallness and a large opening for largeness. Listeners can take it as a
hint for relative size referred to by each sound, i.e. a sound uttered with a
small opening such as a rounded high back vowel /u/ refers to smallness,
although it is one of the back vowels, and an unrounded low back vowel
/a/ has a larger opening and therefore, evokes largeness.
Table 4. Demonstratives in Hausa (Chadic, Nigeria, Wolff 1993: 119-20)

Near speaker Near addressee Away from speaker Further away from
and addressee speaker and
addressee
nân nan cân Can
Figure 2. The amount of closure in the vocal cavity
Bearing in mind this tendency in the sound and its referent, recall the
demonstratives in Somali in Table 1 and Table 3. Contrary to what is
expected according to the general pattern of sound symbolism, e.g. the
English this and that, Somali exhibits an opposite distributional pattern, i.e.
high front vowels are used to refer to distance, and high back vowels to
proximity. Table 5 and Table 6, repeated from Table 1 and Table 3,
respectively, show this exceptional distribution with grey shading. It is
apparent that the demonstrative for far distal behaves according to the
common pattern, and the rest are all exceptions. This type of opposite
combinations is termed here as counter-symbolism, and it is assumed that
a possible motivation can be postulated.
64 Chapter Five
Table 5. Demonstratives in Somali (Saeed 1999: 113)

Proximal Medial Distal Far distal
(close to (further away (in the middle (in the far
speaker) from speaker) distance) distance)
M -kán -káa(s) -kéer -kóo
F -tán -táa(s) -téer -tóo
Table 6. Articles/demonstratives in Somali (Saeed 1999: 112)

Non-remote Remote
M -ka -kii
F -ta -tii
Counter-symbolism into symbolism

In order to decode the counter-symbolism found in Somali, it is argued
here that one has to incorporate the visual perception of a speaker, i.e.
once vision becomes a part of interpretation, counter-symbolism in Somali
ceases to exist.
Reference to size or distance is two-fold, i.e. an actual physical size
and a visually perceived size. A large object, once seen from a distance, is
perceived as a small object. This gap in perception may be a key to
understanding the demonstratives in Somali in terms of sound symbolism,
i.e. what is close to interlocutors appears large or in its normal size to them,
and what is far from them looks smaller. Consider the cows in Figure 3:
they can be more or less the same sizes, but depending on how close they
are to viewers, their physical size appears different. Thus, a physical size
observed by viewers can be an iconic sign to motivate vowel patterns in
Table 5 and Table 6, and the underlying symbolism can be the same as
other common patterns.
A possible reason for this type of the vowel pattern may be related to
the speakers’ traditional nomadic life (cf. Saeed 1999). The Somalis are
traditionally nomadic pastoralists, often dealing with herds of camels or
cattle. They travel through an open land and they can be highly sensitive
to distance in comparison with city dwellers. In other words, their living
environment or culture could have possibly influenced the way the
demonstratives were formed as they are now. This line of argument goes
along ethnosyntax (Enfield 2002), and further research is needed to
identify the link between culture and language in the case of the Somali
demonstratives.
Figure 3. Cows seen at various distances
As far as the case observable in Somali is concerned, vision and

size/distance are related, and what is observable to the naked eye actually
corresponds to the general rules of sound symbolism. However, what
criterion each language chooses between an actual physical size and a
visually perceived size as a base for their sound symbolism has yet to be
clarified. In addition, we also need to observe if this line of argument
holds true and is applicable to other languages, i.e. typological validity,
and whether the distributional pattern, if there is any, can be considered an
areal feature. Areal features are often related to contact, which suggests
that exceptional cases such as the Somali demonstratives could have gone
through areal diffusions, or divergent types of replications (cf. Toyota
2010) due to factors such as social identity, i.e. expression of the self by
not conforming to the areal linguistic norm. These reasons behind
motivations for counter-symbolism may shed light on future research.
Conclusions
This paper has examined some exceptional cases to sound symbolism,
focusing on the demonstratives in Somali and other Cushitic languages.
Somali is grammatically very sensitive to the information structure, with
the presence of the specific focus markers báa and wáxa, i.e. (2).
Similarly, the demonstratives in Somali have also extended its range of
reference from space to the temporal domain and discourse. What is
peculiar in Somali is that its demonstratives do not conform to the general
pattern of sound symbolism, i.e. high front vowels denote smallness or
66 Chapter Five
proximity, and high back vowels, largeness or distance, as shown in Table

5 and Table 6.
This exception has been explained in this paper as follows: an object
seen in a distance looks smaller than a similar-sized object close-by. Thus,
what is referred to by the demonstratives can be varyingly perceived
according to distance. Thus, the distance denoted by the demonstratives is
not reflected in the vowel choice for them, but the choice reflects the size
observable to the naked eye, and what appears to be an exception can be
nicely accommodated within a common understanding of sound
symbolism.
References
Ahlner, F. & J. Zlatev 2011. Cross-modal iconicity: a cognitive semiotic
approach to sound symbolism. Sign System Studies, 38, 298-348.
Carling, E. R. 2004. A grammar of Trio, a Cariban language of Suriname.
Frankrut am Main: Duisburg Papers on Research in Language and
Culture.
Diessel, H. 2013. Distance Contrasts in Demonstratives. In M. S. Dryer &
M. Haspelmath (Eds.), The world atlas of language structures online.
(Available online at http://wals.info/chapter/41, Accessed on 2017-01-
15.)
Enfield, N. (Ed.) 2002. Ethnosyntax: explorations in grammar and culture.
Oxford: Oxford University Press.
Hayward, R. J. & J. I. Saeed 1984. NP focus in Somali and Dirayta: a
comparison of baa and pa. In T. Labahn (Ed.), Proceedings of the
Second International Congress of Somali Studies, University of
Hamburg, August 1-6, 1983, (pp.1-21). Hamburg: Helmut Buske.
Heine, B. & T. Kuteva 2002. World lexicon of grammaticalization.
Ikegami, T. & J. Zlatev 2007. From non-representational cognition to
language. In T. Ziemke, J. Zlatev, R. M. Frank (Eds.), Body, language
and mind, Vol 1: embodiment, (pp. 241-283). Berlin: Mouton.
Johansson, N. 2011. Motivations for sound symbolism in spatial deixis: a
study of 101 languages. BA thesis, Centre for languages and literature,
Lund University.
Muysken, P. 2008. Nominal tense. Time for further Whorfian adventures?
Commentary on Casasanto. Language Learning, 58, 81-88.
Nordlinger, R. and L. Sadler 2004. Nominal tense in crosslinguistic
perspective. Language, 80, 776-806.
Oomen, A. 1978. Focus in the Rendille clause. Studies in African

Languages, 9, 35-65.
Saeed, J. I. 1999. Somali. Amsterdam: John Benjamins.
Sasse, H-J. 1981. “Basic word order” and functional sentence perspective
in Boni. Folia Linguistica XV/3-4, 253-290.
Toyota, J. 2010. Identity-related issues in contact-induced historical
changes. In Lopiþiü, V. and B. Mišiü-Iliü (Eds.), Language, Literature
and Identity, (pp. 127-138). Newcastle upon Tyne: Cambridge
Scholars Publishing.
Traunmüller, H. 1994. Sound symbolism in deictic words. In H. Auli & P.
af Trampe (Eds.), Tongues and texts unlimited. Studies in honour of
Tore Jansson on the occasion of his sixtieth anniversary, (pp. 213-234).
Dept. of Classical Languages, Stockholm University.
Wolff, H. E. 1993. Referenzgrammatik des Hausa: Zur Begleitung des
Fremdsprachenunterrichts und zur Einführung in das Selbststudium.
Hamburg: LIT Verlag.
Zlatev, J. and Andrén, M. 2009 Stages and transitions in children’s
semiotic development. In J. Zlatev, M. Andrén, M. Johansson Falck &
C. Lundmark (Eds.), Studies in language and cognition, (pp. 380-401).
Newcastle: Cambridge Scholars.
CHAPTER SIX
COMPARATIVE ANALYSIS OF ANCIENT

CHINESE CHARACTERS AND MODERN
PICTOGRAMS
DARIA VINOGRADOVA
Abstract. The number of visual symbols we use in communication and

information acquisition is significantly large: emoji 1 in digital
communication, pictograms in public places, pictorial symbols for
technical devices, and road signs. In the 20th-21st centuries, pictograms
became a part of our everyday life more than ever before, but it does not
mean they are new. In this paper, we will discuss similarities and
differences between ancient Chinese characters and modern pictograms,
focusing on structural and compositional features of these two groups of
symbols. First, we will examine the role of the position of individual
elements in the whole structure of symbols. Next, we will focus on the
emergence of new meanings through the transformation of pictograms and
characters’ components. We will explore the possibilities of producing new
symbols by transformation of basic elements in ancient pictograms. Also,
we will analyze the methods of depicting movement in static symbols as a
relatively complicated and elaborate way of representing objects.
Introduction
In 2015 the ‘Face with tears of joy’ emoji was chosen by the Oxford
Dictionaries, a blog site run by the Oxford University Press, as the Word
of the Year. For the first time in history not an actual word but a pictogram
became the one reflecting the mood of the past year. In modern society,
different types of pictograms play a significant role in everyday life. We
use emoji in digital communication, we get information from pictograms
1
The word “emoji” comes from Japanese, and means pictograms used in digital
communication.
70 Chapter Six
in public places and operate devices using pictorial symbols. In the era of
globalization, when people from different countries travel the world
physically and virtually, pictograms have become some sort of universal
language playing a significant role in communication and information
acquisition.
Crow (2006) points out the following advantages of the pictorial word:
“The ability of images to communicate across linguistic boundaries offers
a level of consistency that is difficult to achieve otherwise. It also has
distinct cost advantages. In a global economy, the ability to distribute the
same product in a number of territories saves both time and money” (Crow
2006:19-20). In the 20th-21st centuries, pictograms became a part of our
everyday life more than ever before, but it does not mean they are new.
“Cave paintings preceded written language by hundreds of centuries, and
the alphabet evolved slowly from earlier pictographic scripts. Ironically,
the development of computers has led to a rediscovery and redeployment
of this prehistoric form of communication. The use of icons is both ancient
and current” (Horton 1994: 1).
If we compare modern pictograms with the ancient ones, we will find a
lot of similarities in the ways of depicting the world around us with a
limited amount of visual techniques. Table 1 demonstrates some examples
of these similarities. We can find simple symbols like ‘horse’ or ‘vehicle’,
depicting the object similarly, and more complicated symbols representing
actions like ‘to cry’ or ‘to take a shower’. Ancient characters like ‘to take a
shower’ are more restricted by graphic tools and have some specific
features making them visually less obvious than modern pictograms, but
close observation of it reveals a person standing in a vessel and water
being poured on him, which is almost the same as the modern form.
Table 1. Similarities in ancient characters and modern pictograms
Meaning Modern pictograms Ancient Chinese characters
car, vehicle

horse
cry
walk

take a shower

Analysis of Ancient Chinese Characters and Modern Pictograms 71
Brief observation of these examples showed us some similarities

between the two groups of visual symbols. We will discuss it more
thoroughly further.
The reasons for examining ancient characters were pointed out by
Horton (1994): “As designers, it provides us with thousands of tested
examples to draw upon. We can observe what has and has not worked
over the centuries. It also benefits us by ensuring us that the users of our
systems have already had considerable experience interpreting visual
symbols” (Horton, 1994: 7). In this paper we will compare the methods of
pictorial representation, examine structure and spatial composition in both
groups of symbols and bring to light the characteristics of modern
pictograms that cannot be found in ancient ones. By modern pictograms
we mean a wide range of different types of visual symbols we use in
communication and information acquisition: road signs, technical
instructions on products, pictograms in public places, computer icons, and
emoji to name a few. We will use the database of visual symbols “The
noun project”2 with a current total number of about 500,000 pictograms.
All pictograms in this database are monochromic and that makes
comparison with monochromic Chinese characters more equal. In Ancient
Chinese characters we will focus on the oracle-bone script, the oldest
Chinese characters depicted on animal bones and turtle plastrons, and
Chinese bronze inscriptions as the oldest forms of characters with a high
level of visual informativity.
Structural and compositional similarities in ancient

Chinese characters and modern pictograms
In this section, we will discuss similarities and differences in structure and
spatial composition between ancient Chinese characters and modern
pictograms. At first glance, there is a lot in common between the two
groups of visual symbols. Both types of symbols consist of a small number
of elements depicting the object focusing on the most significant features
of it. There are a lot of possible ways to compare two types of characters
according to their structure. We will focus on three of them.
First, to examine a role of the position of individual elements in the
whole structure of a symbol we will analyze examples consisting of the
same elements. The way these elements are positioned defines the
meaning of the character. For example, in modern pictograms ‘going up
2
“The noun project” is a website (thenounproject.com), founded in 2010 as a
recourse for creation and sharing symbols used in digital communication.
72 Chapter Six
by escalator’ and ‘going down by escalator’ there are three elements:

escalator, person and arrow. When we change the direction of the arrow
the meaning changes to the opposite. We assume that the same rule works
in ancient characters, too.
Next, we will focus on the emergence of new meanings through the
transformation of pictogram and character components. A lot of symbols
are made of a small amount of visual forms. In pictograms, a simple
change of the pose of the basic element ‘person’ 3 creates a variety of
4
new meanings, like ‘fall’ or ‘walk’ 5 . We will explore the
possibilities of producing new symbols by the transformation of basic
elements in ancient pictograms.
Finally, we will compare one of the ways to perform complicated
meanings. We will apply the classification of different methods to depict
movement in static pictures developed and reproduced by Murayama
(1988) on the base of principles proposed by Ward (1979), Friedman &
Stevenson (1980).
Spatial composition
In this part we will focus on spatial composition in visual symbols and its
correlation with meaning. The most obvious way to prove the crucial role
of spatial composition of elements is to observe the characters with the
same or a similar set of elements. Ancient characters ‘top’ and ‘bottom’
are a good example of this feature, and the simplest one. Both consist
of short and long lines. If the short line is above the long one, it is ‘top’, if
it is under, it is ‘bottom’. We can find the same method in more
complicated characters too for example, ‘together’ (ඹ) and ‘compete,
fight’ (த). Modern forms of these characters don’t have much in common,
but the ancient ones consist of almost the same set of elements, positioned
differently.
The ancient form of the character ‘together’ consists of two hands
holding an object. Both hands are drawn symmetrically, showing equal
effort and power. Henshall (1998) gives the following explanation of this
character 6 : “Two hands offering a jewel. Focus on the idea of doing
3
https://thenounproject.com/search/?q=person&i=49525
4
5
6
The reason we use “A guide to remembering Japanese characters” by Henshall
for English explanations of characters is it is based on the most conventional
meanings and authoritative theories.
something with both hands and by extension jointly/together.” (Henshall

1998:529) The ancient form of the character ‘compete’, ‘fight’ also
consists of two hands and object. Henshall explains this character “shows
a hand reaching down and another hand holding an item, for possession of
which the hands are presumably vying” (Henshall 1998: 163). Both these
characters consist of three elements, i.e. Hand 㸩 Hand 㸩 Object. The
horizontal and symmetrical position of hands shows stability and equal
power, however changing the position to diagonal gives an effect of
tension and dynamism between the hands.
Modern pictograms ‘handle with care’ and ‘exchange’ have a
similar structure. The first pictogram depicts two hands holding an object
and is almost similar to the ancient character. The second pictogram shows
one hand giving an object to another. It is worth pointing out that in
modern pictograms movement is often depicted horizontally (Horton
1994) rather than diagonally as we can see in ancient pictograms. Both
ancient characters and modern pictograms have the same structural
formula A㸩A㸩B, i.e. Hand㸩Hand㸩Object. B may slightly differ, but
the relocation of A produces new meaning.
As mentioned above, changing the direction of an arrow in modern
pictograms and makes the opposite meaning. In ancient Chinese
characters one of the most frequent elements ‘footstep’ may play the
role of an arrow too, showing direction. For example, the character ‘to
exit’ looks like a foot exiting some square form, a house presumably.
Changing the direction of this element gives a new meaning ‘place’ . If
the footsteps are positioned in opposite directions it is the character
‘different” , if crossing a wavy line it is ‘to cross” . The same
composition can be found in modern pictograms , , .
The importance of spatial composition is one of the key features in
ancient characters, but it almost was not inherited by modern characters, as
demonstrated in Table 2. In this table, we compare ancient and modern
forms of the characters, including the element ‘water’. In the first four
characters the element ‘water’ is depicted in the form of a river. For
example, the character ‘to sink’ shows an ox sinking in a river and the
interaction of the two elements is clear. The last two characters depict
water as spit and tears. For example, the character ‘to cry’ depicts an eye
and tears falling from it.
Interaction of the elements in ancient characters disappears in modern
ones. In all six characters water is positioned on the left, another element
on the right and there is no visual connection between them.
In this part we have compared ancient Chinese characters and modern
pictograms and confirmed the similarities between them in the field of
74 Chapter Six
spatial composition. Next, we will focus on the forms of the structural

elements.
Table 2. Differences in spatial composition between ancient and modern

characters
Ancient Modern Meaning/structure (Mizukami

character character 1995)
1. ỿ Ox sinks in water
2. ⁠ Grass and trees in river basin
3. ΅ To cross a river
4. To fish in a river
5. Open mouth waters
6. Ἵ Eye and tears
Transformation of basic elements

Spatial composition as well as the transformation of basic elements
produces a variety of new meanings. In ancient characters the basic form
‘person’ with an enlarged head means ‘elder brother’ , and seating a
person with hands doubled in length and long fingers means ‘to take’ .
In this part we will focus on the basic forms ‘person’ and parts of the
human body and reveal different ways to create new visual symbols by the
transformation of the basic ones. As the number of ancient characters is
relatively limited, especially in comparison with modern pictograms, first
we will focus on the ancient characters and attempt to find the same
features in modern pictograms.
One of the simplest ways to draw attention to something is to depict it
bigger than other parts or to make it more detailed. In ancient characters,
enlarging or detailed depiction of a part of a body draws attention to this
part and thereby names it. For example, the ancient character
‘buttocks’ consists of simple basic form ‘person’ with enlarged buttocks
and the character ‘knees’ shows a person with enlarged knees . In
modern pictograms a more frequently appearing way to show a part of a
body is to draw just this part, to depict it in more detail or in color in
contrast with the remaining part, like in the pictogram ‘kidney’ . Also in
ancient characters an exaggerated depiction of one part of a body is used
as metonymy and represents some function. For example, the character ‘to
see’ draws a seated person with a big eye instead of a head . In modern
pictograms the depiction of eyes is more frequent for this meaning.
One more way of transformation of the basic element is adding another
basic element and interaction with it. In ancient Chinese characters and
modern pictograms we can find similar symbols with this feature. The
Chinese character ‘old’ depicts an old person with a walking stick . The
modern pictogram looks almost the same .
A final feature shows a significant level of figurativity in both groups
of symbols. A part of a body is replaced by some object. In the Chinese
character ‘fire’ it is a head replaced by fire. In a modern pictogram it is
a human brain replaced by gears .7
Representation of movement
In this part we will compare ancient characters and modern pictograms
focusing on the way they perform complicated meanings. We will apply
the classification of different methods to depict movement in static
pictures proposed by Ward (1979), Friedman & Stevenson (1980) and
developed and reproduced by Murayama (1988). Murayama analyzed
pictures from different cultures, periods and artists from cave paintings to
modern artists and examined the way movement was indicated.
The indicators Murayama proposed on the base of previous researches
by Ward and Friedman & Stevenson are as follows:
1) Posture: “Representation of the body parts so they appear to be

engaged in a coherent stage of an action, such as generating motion
or preparing to strike the ground, so that the interacting roles of
gravity and momentum, both present and anticipated, are clear.”
(Ward 1979: 255)
2) Context: “Context, in which objects are depicted may suggest
movement: A skier with skis pointed down a hill must be moving.”
(Friedman & Stevenson 1980: 227)
3) Interaction: “Subjects that show energy transfer may improve the
coherence of the action. Also, the amount of pictorial space shown
before or behind a figure in forward motion may affect perception.”
(Ward 1979: 257)
7
https://thenounproject.com/search/?q=think&i=103063
76 Chapter Six
4) Contingency: “Tendency of flexible parts, such as hair and clothing,

to trail behind and specify both the direction and speed of the
movement.” (Ward 1979: 256)
5) Necessitating: “Some objects and scenes inherently suggest
movement by their content. For example, there is no such thing as a
still river or fire. Movement, therefore, is inferred in pictures of these
subjects without additional indicators - the content forces the
interpretation.” (Friedman & Stevenson 1980: 227)
6) Metaphor: “Aspects of the environment that are unlikely to occur
together in the real world may be represented side by side in a picture
so as to suggest movement.” (Friedman & Stevenson 1980: 227)
7) Abstraction: “Pictorial movement can be indicated without any
pictorial object … angular and curved lines convey a strong
impression of movement.” (Friedman & Stevenson 1980: 227)
8) Trajectory: “Path an object was taken. Footprints on the sand, ski
tracks on fresh snow, skate marks on ice, wheel marks in mud, dust
kicked up by a horse, the wake behind a swimmer or a ship, or
arrows next to an object.” (Friedman & Stevenson 1980: 228)
9) Multiple viewpoints: “To show an object or a part of an object at
successive moments. In this way, a selection of the changes occurring
with time is recorded as multiple images.” (Friedman & Stevenson
1980: 227)
We have examined ancient characters and modern pictograms, and

determined the indicators representing movement, as summarized in Table
3. It shows the frequency of every type of movement indicator in ancient
Chinese characters and modern pictograms and typical examples, with
similar meanings for both groups when possible. All indicators were found
in groups of symbols, and some similarities and differences in particular
groups were detected. Due to the limited number of ancient characters and
almost unlimited and growing number of modern pictograms it is
impossible to compare the total number of indicator usage frequency
between the two groups of symbols. Therefore, we can evaluate the ratio
of one particular indicator by comparing it with the others in one group.
Posture, context and interaction are the most frequent indicators in
both groups of symbols. Previous research (Murayama 1982; Friedman &
Stevenson 1980) shows the same results for most of the pictures from
different cultures, periods and styles. Ancient characters show a smaller
number of examples for contingency, abstraction and multiple viewpoints
presumably due to a relatively realistic way of object observation and
depiction.
Table 3. Pictorial movement indicators in ancient characters and modern

pictograms
Pictorial Ancient Number Modern Number of

movement characters of cases pictograms/ cases
indicators /meaning meaning
Posture: 46 96

‘incline’ ‘run’
Context 47 83

‘fight’
‘cut’
Interaction 36 67

‘fight’ ‘fight’
Contingency 5 13

‘middle’
‘flag’
Necessitating 10 19

‘river’
‘river’
Metaphor 11 27

‘rain’ ‘rain’
Abstraction 4 32

‘hurricane’
‘cloud’
Trajectory 23 30

‘follow’ ‘run’
Multiple 1 10
viewpoints ‘move’
‘lick’
In this part we have analyzed structure, spatial composition and

movement representation in ancient Chinese characters and modern
pictograms. Color and animation are frequently used in modern
pictograms, graphic tools are indisputably elaborate and the number of
images modern people see and operate is incomparable with the ancient
ones. Nevertheless, a significant amount of similarities in compositional
features and the ways objects are represented can be detected between
ancient Chinese characters and modern pictograms.
78 Chapter Six
Conclusions
In this paper we have analysed some structural and compositional features
of ancient and modern pictograms and detected a variety of similarities in
the basic principles of formation of both types of symbols. There is a lot in
common between the way ancient and modern people observe and
reproduce the world around them in the form of visual symbols. Similar
principles have been detected in the spatial composition of both groups of
symbols and we proved that the way elements of a visual symbol are
positioned defines the meaning of the symbol. Also, we have detected that
in both groups a lot of symbols are created of a small number of basic
forms and their transformations. However, the number of visual symbols
used in modern pictograms is incomparable with ancient characters. It
contains all visual symbols accumulated in human history and the Chinese
ancient characters are a part of it. Finally, we have tested the way
movement is depicted in ancient characters and modern pictograms.
Showing movement through posture, contest and interaction is equally
frequent for both groups of symbols, but modern pictograms demonstrate
significantly more examples for contingency, abstraction and multiple
viewpoints.
Modern pictograms have a lot in common with ancient characters.
Nevertheless, significant differences can be found in the way modern
pictograms appear and spread and the role they play in communication.
The speed of transmission of new pictograms is incomparable with any
other types of pictograms that ever existed. Almost everybody can create a
new pictogram and share it. As mentioned before the current number of
pictograms in the database of “The noun project” is half a million. To
create a pictogram worth being uploaded to the database a creator must
follow some simple technical guidelines and principles of design. There is
almost no limitation in themes and the number of pictograms. Also,
modern pictograms being a visual language are not pretending to
substitute for natural language. Despite the significant amount of
meanings performed by pictograms, there is still a considerable part that
can be expressed only by natural language, which is concrete and
relatively unambiguous.
References
Crow, D. 2006. Left to Right: The Cultural Shift from Words to Pictures.
AVA Publishing.
Easterby, R., Zwaga, H., eds. 1978. Information design. The design of
evaluation of signs and printed material. New York: Wiley.
Friedman, S.L., Stevenson, M.B. 1980. Perception of movement in
pictures. In M. A. Hagen (Ed.), The Perception of pictures, (pp. 225-
255). New York: Academic Press.
Henshall, K. G., 1995. A Guide to Remembering Japanese Characters.
Tokyo: Tuttle Publishing.
Horton, W. 1994. The icon book: visual symbols for computer systems and
documentation. New York: Wiley.
Li, Zong Kun. 2012 Jia gu wen bian (Compilation of oracle bone scripts).
Beijing: zhong hua shu ju.
Mizukami, S., 1995 Kokotsubun kinbun jiten (Dictionary of oracle-bone
and bronze scripts). Tokyo: Yuzankaku.
Murayama, K. 1988 Shikaku geijutu no shinrigaku (Psychology of visual
art). Tokyo: Seishinshobo.
Ota, Y. 1993 Pikutoguramu “emoji” dezain (Pictogram design). Tokyo:
Kashiwashobo.
Ota, Y., Kato, H., Sato, T., Nakagomi, T., Murakoshi, A., 1983 Me de
miru kotoba no sekai (Visual world of languages). Japan: Nihon
kikaku kyokai.
Ward J. L., 1979. A piece of the action: moving figures in still pictures. In
C. F. Nodine & D. F. Fisher (Eds.), Perception and pictorial
representation, (pp. 246-271). New York: Praeger.
Xu Zhongshu. 1988 Jia gu wen Zidian (A dictionary of Oracle-bone
Inscriptions). Chengdu: Sichuan cishu chubanshe.
The noun project homepage. (http://www.thenounproject.com) (accessed
on 2 October 2016)
PART THREE:
ANTHROPOLOGICAL
AND LINGUISTIC STUDIES
CHAPTER SEVEN
VISION IN AFRICAN LANGUAGES
MARILENA STÜWE-THANASOULA
Abstract. This chapter presents an overview of the little studied area of

vision in African languages. The fragmentary documentation and analysis
on this subject mirror the lack of information. What are available to us now
include the polysemy of visual verbs, evidentials based on visual concepts,
the notion of the invisible in African cultures and the use of the visual body
part ‘eye’ within the domain of social interaction and emotion. These
points are illustrated to shed light on the intricate nature of language and
culture in Africa concerning vision.
Introduction
When it comes to African languages a very common comment and the
most common finding irrespective of the concrete scientific matter is that
more research is needed. The following chapter will present a few notes on
vision in African languages and thus I begin with this statement, since
very little is published on perception in African languages in general and
on the linguistic aspects of vision in particular (for an overview of
publications on related matters in African languages see Thanasoula 2016:
68-70).
This chapter gives an overview of recent publications in the field of
vision in African languages and will be organised as follows: first, I will
briefly sketch some issues of the scientific debate within cognitive
linguistics that gave the impulse to a number of publications on language
and perception in African languages. Second, I will provide some
information on the ethnology of the body and will present some cultural
concepts that help us understand the meanings linked to vision from the
perspective of various African cultures. Then I will give an overview of
the domain vision in African languages, referring to recent linguistic work.
Instead of an outcome this chapter will end with some thoughts and open
questions related to the way perception is expressed in African languages.
84 Chapter Seven
A note on methodology here: recent linguistic research on perception,

including vision, in African languages is raised in the course of cognitive
linguistics as well as within the linguistic debate concerning possible
universal semantic and/or structural features that are related to the way
human language expresses categories of perception. Various authors use
various methods and theories to approach the matter and it is not possible
here to provide enough information on the partly very different
methodological and theoretical background of each work. The aim of this
chapter is to sum up and present some key issues concerning vision in
African languages and not a comparison of methods and theories on
perception as they are applied in African linguistics.
The matter of translation comes up as one of the core challenges,
especially when it comes to language and perception: both the choice of
translation for the explanation of an utterance made in a language of
interest as well as the metalanguage, which science develops to analyse
phenomena, impact considerably on the collection and the analysis of
linguistic material.. However the choice of words may be considered as the
light part of the problem, when it comes to the translation of perception.
The core challenge is to identify, describe and explain adequately senses
and sensory categories, which are not recognized as such within Western
culture. Despite numerous publications on senses beyond the five
evaluated as basic ones (because they are linked to a basic sensory organ),
science in general, and linguistics as part of it, are captured in a viewing
culture. Consider for example that visualization is evaluated as very
important in the fields of technical innovation (Van Beek 2010: 249).
Perception remains a domain between nature and culture and science
raises a claim of objectivity, which is difficult to keep up with (Cf. Fabian
1994 on vision and observation within science).
I would like to give an example of a sense that is not included in the
Western concept of what is supposed to be a sense from my own
fieldwork. The word amakulu in the Eastern Bantu language Lushese, as
demonstrated in (1), expresses the wholeness of the human body
paradigmatically: it represents both the meanings ‘sense’ and/or ‘knowing’
and is interpreted as a main human capacity. The meaning of the word
amakulu expresses precisely that there is no knowing without a sensual
experience and implies that every sensual experience is a source of
knowing for human beings.
The sense amakulu is what makes the difference within the world of
animates between human and non-human entities from the point of view
of the speakers of Lushese: it is considered as a basic capacity of the
Vision in African Languages 85
human body, for which the whole body is conceived as the organ of
perception.
Lushese (Thanasoula 2016: 204) 1

(1) O-bwá-kintu ba-lína a-ma-kulu,
ART-14REL-Kintu 3PL.PRS-have ART-6-sense/meaning
o-bwa-irála te-ba-lina.
ART-14REL- NEG-3PL-have
‘Human (beings) have amakulu, other (beings) don’t.’
Is it possible to speak about the human body in an objective way? The

answers differ considerably among scientists and among various scientific
disciplines. The rise of cognitive science in general as well as of its
linguistic branch in particular gave impulse to vehement oppositions
concerning the way nature and culture impact physical, emotional,
cognitive and social phenomena. The linguistic debates include issues on
how language is used to express physical experience and further
perception as well as metaphors associated with the body and body parts.
Further the relationship between body and mind and between perception
and intellect have been the target of research especially within cognitive
linguistics.
The predominance of vision over the other senses, which constitutes a
crucial point of many models dealing with the issue of language and
perception, is challenged mainly by anthropological and descriptive
linguistic studies, 2 whose authors criticize the claim that cognition is
universally linked to vision as being Eurocentric (for a summary of the
linguistic debate see Aikhenvald & Storch 2013 and Thanasoula 2016
among others). The evaluation of the senses turns out to be a flexible and
multidimensional matter due to its dependence on various social, cultural
and personal parameters in interaction (see, e.g. Belting 2008). Still, the
hegemony of vision developed within Western culture dominates the
evaluation of the senses in scientific discourse. Looking for the basic sense,
science rehashes the answer, and since vision is assumed to be the basic
sense, the whole visual domain gains importance. The dominance of the
1
Note that Bantu languages have nominal classes, which are glossed through plain
numbers. It is worth pointing out that the function of the morpheme, which is here
glossed as ART, is still a matter of controversy. For an overview on the debate, see
Thanasoula (2016: 84-87).
2
For critical reviews from a typological point of view on language and perception
based on cross-linguistic comparison, see various publications by the Max Planck
Institute for Psycho-linguistics.
86 Chapter Seven
scientific image on vision and its presumed direct relation to cognition

results in economic and political aspects concerning research in Western
institutions of higher education which cannot be discussed here in this
limited space (for the interrelation between science and society see Young
1987 among others). As a kind of interim balance sheet on the issue of
universals concerning language and perception, Aikhenvald & Storch
point out that “one thing is clear at our present state of knowledge: no
‘hierarchy’ of senses is universal” (Aikenvald & Storch 2013: 37, for more
details see Evans & Wilkins 2000 and Viberg 2001).
Every sensory domain is associated with specific cultural interpretations
as well as practices. In this sense, all domains of perception have their own
unique semantics and thus importance. The specific interpretation and
evaluation of properties should be more integrated in the analysis of
sensory modalities and their linguistic expression, since to recognize
properties as such is a matter of cultural experience; further, to evaluate
them and to communicate about these through language are matters of
socialization. An interactive approach to the relation between nature and
culture implies that “whatever we are and do always bears some imprint of
our corporeal existence plus just as inevitably of our culture, and that
interrelation is always dynamic” (van Beek 2010: 245).
The terms cultural conceptualization and cultural metaphor (Lakoff &
Johnson 1980, Lakoff & Turner 1989 and Kövecses 2002 among others)
are developed to allow alternative views on how language develops
meaning within specific culturally-defined discourses and how culture-
dependent images impact the choice, the use and the meaning of linguistic
means.
Before I turn to the cultural conceptualizations of vision as attested in
African languages a note on the distinction between scientific and non-
scientific approaches is appropriate: how people in general talk about an
issue and the scientific analysis of the same subject constitute two
different ways to approach any matter. A seldom referred-to and therefore
precious example of the relationship between expert theories and folk
theories is presented by Kövecses in his work on the metaphors of emotion.
He investigates the relationship between folk theories on emotion based on
ordinary language and scientific theories that are constructed to account
for the same issue and observes that the assumed discrepancy between folk
theories and expert theories is oversimplifying a rather complex issue. He
points out four aspects of the relationship between folk understanding and
scientific theories: a) that scientific theories on emotion often elaborate at
least one aspect of the folk model; b) that there is a positive correlation
between the acceptance of scientific theories and the amount they overlap
with folk models; c) that scientific models that explain emotions in terms
of concepts missing from the folk understanding of emotion may appear
more scientific but are less appealing in matters of intuition; and d) that
scientific theories spread ordinary but mistaken beliefs of folk models
(Kövecses 2000: 126-127). The author suggests that the investigation of
the relationship between folk models and the scientific approach should be
elaborated more into a scientific analysis and include historical parameters
in order to understand the development, the differences and similarities of
emotion concepts (ibid: 138). In that sense I will turn now to some
concepts that are often described in various works on African cultures and
languages, which are concerned with the body and perception in general,
and based on this I will enlighten the parameters that play a role in
understanding the domain of vision from the perspective of some African
cultures.
Cultural conceptualisations concerning the human body

and human perception in various African cultures
As mentioned in the introduction more research is needed in order to
understand the way categories of perception are expressed in African
languages. The same remark holds true for the cultural conceptualizations
concerning the body and perception in general as well as vision in
particular in African cultures. Not only is very little said about these
matters, but further the vast number of cultures and languages of the
African continent causes a significant challenge to start with: is there
anything that one could depict as common among the various African
cultures, when it comes to the interpretation of the body and its senses?
The answer is no, at least for now, since more data are needed. Still there
are some generalizing remarks in numerous studies on the ethnology of the
body, which I will here put together. The notion of personhood and
metaphysical concepts, as well as the high evaluation of social interaction,
seem to be salient aspects for understanding the domain of perception in
various African cultures.
Starting with the interpretation of the body in many African cultures
and languages three aspects are broadly attested: a) a holistic image of the
body, b) the image of the body as a container, and c) the image of society
and/or the whole creature as a body. The holistic approach to the human
body implicates that the distinction between the physical, the emotional
and the mental domains does not exist from the point of view of many
African cultures. The human body is regarded as a unity characterized by
various capacities and qualities that cannot be isolated, separated from
88 Chapter Seven
each other or evaluated in comparison. From the perspective of various

African cultures the human body, the social body and nature share
spiritual and material elements: they are made of the same substances. To
put it in the words of an old man on the Ssesse Islands in Uganda “All is
one, don’t you see, when the forest suffers, the lake is ill, when the brain
suffers, the body is lame and when the body suffers you can’t concentrate
on anything else but the pain” (Sseluwagi Dominic, May 2009 during
fieldwork in Thanasoula 2016).
The conceptualization of the body as a container has been described in
many cultures around the world beyond African ones (Cf. Kövecses
2002). Describing the body conception within the Swahili culture of
Mombasa, Swartz mentions four elements, namely cold, hot, dry and wet,
and comments as follows: “The same four elements are present in everyone,
but there are important differences among individuals in the relative
amounts or strength of each and in the nature of the balance among the
four of them (Swartz 1992: 41)” (Thanasoula 2016: 205).
The access to the own self, the access to other humans as well as to
other living organisms and the surroundings in general are possible only
through the body. Behrend underlines its significance in the pre-colonial
cosmology among the Tooro in Western Uganda: “in the give and take of
social life, connections that emerge between individuals are embodied
connections in the fullest sense. It is from the materiality of the body, its
flesh, blood, milk, semen, spittle and so forth that relationships are
nurtured, formed and transformed” (Behrend 2011: 4).
The exchange and transformation of substances result, among other
processes, out of actions of incorporation: “The idea of incorporation
depends upon and creates a radical division between inside and outside, an
inside often associated with good while the outside tends to be related to
bad. It is this division that produces the desire to return to oneness and
total unity (Kilgour 1990: 4-5)” as cited in Behrend (2011: 27). Besides
preparing, sharing and consuming food, other forms of incorporation
include sex, spirit possession, and verbal communication as well as
performance and ritual experience. All these domains of incorporative
action are accessed through the human body and at the same time they
define and transform the position of the individual within the social and
cosmic frame through the corporal experience.
The concept of an open body which can be compared to a container
capable of accessing to and exchanging of various spiritual and material
elements is attested in various African languages. Furthermore, the notion
and importance of balance must be underlined: while the balance of
elements, which may fill the body-container, is positively associated, the
lack of balance is consequently interpreted in negative ways. Suffering, for

example, is often conceived as losing balance and/or control in various
African cultures: further, conditions of suffering are not regarded as being
personal or individual: social parameters build the internal causal frame of
a disease, like the respect for norms and prohibitions or the relations
between members in the social group where the person fell ill. In general
the intentions and behaviour of the social group and each of its members
invite external forces and allow them to enter the body of a member and
through it the body of the community: “A number of studies of
disturbance and illness in Africa have equally stressed that whilst
causation of misfortune may be looked for externally, this is usually
accompanied by strong feelings of guilt, moral anxiety and disturbances of
conscience” (Jackson 1981 among others). “An important feature
emerging from these studies is the rationalisation of the causes of illness
as due to actions of a malevolent outside force, the reasons for its
successful invasion being seen to be due to the disharmony and confusion
caused by disputes, bitterness and envy within a social group” (Rowlands
1992: 120).
The concept of suffering not only links physical, emotional and cognitive
conditions but it builds a bridge to the metaphysical world: the context of
spirits’ possession and witchcraft also includes loss of balance and control;
it is referred to as illness and it is associated with invasion of the body by
external forces. Still these forces are not evaluated always as malevolent:
spirits may affect humans positively, even rescue them. The notion of
agency is more ambivalent compared to the loss of control in the case of
purely physical suffering, because in a case of spirits’ possession the
initial loss of control may lead to exceptional power and knowledge.
The holistic image of the creature and of the human body and the
flexibility and interrelation of the elements constituting the world impact
the local concept of perception as described for African cultures. The
importance of social interaction as well as the connection of physical with
metaphysical entities as parts of the body of society impact considerably
on the image of the senses and of sensory categories and further physical,
emotional and cognitive experience. To approach perception from the local
point of view of various African cultures presupposes an understanding of the
domain of social interaction.
Various publications on the way African languages express perception
stress the predominance of the auditory domain, because it shows the most
inter- and intrafield polysemies and includes semantic links to the domains
of emotion and cognition. It is basically the sense of hearing that is linked
to cognition and to knowledge as attested in various African languages.
90 Chapter Seven
While in Western culture and science visual information is evaluated

higher than auditory evidence, this is not the case in numerous cultures in
Africa, where the weight of auditory evidence is evaluated as the highest.
Often the verb which expresses auditory experience and corresponds to the
English term ‘hear’ has a holistic meaning which may be translated in
English as ‘feel, perceive, sense’ and ‘understand’. The domain of vision
on the other hand shows fewer links to other domains of experience. Van
Beek points out that “hearing versus seeing is a contrast of cognitive
styles, and the appreciation of the senses is tied to appreciation by the
senses, the importance of seeing is linked to the importance of what to
see” (Van Beek 2010: 255). The culture-based logic of the senses impacts
the weight of the senses and consequently the weight of vision. Still the
language of vision allows precious insights into concepts of perception
and social interaction from the point of view of African cultures.
The domain of vision in African languages

It won’t be possible here to define the domain of vision because of space
limitation. I will focus on the polysemy and grammaticalization of the
visual verbs, and on words used to refer to visible and invisible categories
as well as on the cultural interpretation of the eye. While the colour terms
as well as the symbolic interpretation of colours should engage us, colours
will be excluded here, because the fragmentary character of the data on
colours as they are expressed in African languages does not allow at the
moment any general comment on colour terms in African languages.
Visual verbs in African languages

Starting with the intra- and interfiled polysemies of visual verbs in African
languages, most authors stress the predominance of the auditory-holistic
verbs, as mentioned above. Metaphors based on visual verbs are motivated
mainly by the image of people meeting face to face, hence the verbs of
vision in African languages are often used within the domain of social
interaction. The following examples in Dongolawi (Nilo-Saharan, Soudan)
illustrate this paradigmatically.3 The verb nal expresses visual activity in
Dongolawi and may be translated with the English terms ‘see; look at;
look for; watch’, as in (2a). Jakobi & El-Guzuuli explain that the semantic
extension of the visual verb into the field of social interaction produces the
3
In order to make the reading of these examples more readily available, the nasal
vowels and some tones are here simplified.
meanings ‘greet; meet; visit; look after; guard’ and ‘protect’, as

exemplified in (2b) to (2e). 4
Dongolawi (Jakobi & El-Guzuuli 2013: 200-201)

(2) a. Booliis magas-ki dukkaan-do too-buu-n nal-ko-n.
police thief-OBJ shop-LOC enter-PROG-3SG see-PRF-3SG
‘The policeman saw/watched the thief enter the shop.’
b. Ay-gi nal-os
1SG-OBJ see-ASP1
‘Greet me/say hello to me/shake hands with me!’
c. Ay Esmaan-gi suug-ir nal-kori.
1SG Osmaan-OBJ market-LOC see-PRF.1SG
‘I have seen/met Osmaan in the market.’
d. Ay wide taa-ri bokkon in an
1SG return come-PRS.1SG until this my
bitaan-gi nal
child-OBJ see.IMP.2SG
‘Look after/guard my child until I come back.’
e. Ek-ki ARTi nal-ko-n.
2SG-OBJ God protect-PRF.3SG
‘God has protected you.’
The use of a visual verb to express the meaning of taking care, looking
after somebody is common in languages in and beyond Africa. A
knowledge of social hierarchies as well as extralinguistic knowledge of the
world (Weltwissen) have a considerable impact on the interpretation of
linguistic expressions. The following examples in (3), taken from Lushese
(Bantu, Uganda) illustrate a different tendency to interpret the verb boina
‘see’ in Lushese by evaluating the social hierarchy between subject and
object.5
The tendency of the speakers to interpret the verb bóina ‘see’ in a
different way means that another interpretation is possible, for example
that children may look after elders, but this would happen under certain
circumstances and thus it is evaluated as less probable.
4
Note that in (2e), the original interlinear gloss of the verb nal was ‘protect’
(Jakobi & El-Guzuuli 2013: 201).
5
These examples are from fieldwork notes, conducted by Thanasoula in 2010.
Consider also the same phenomenon concerning the interpretation of the auditory
verb in Thanasoula (2016: 333).
92 Chapter Seven
Lushese (Thanasoula, unpublished fieldwork material)

(3) a. A-ba-kairè ba-boína o-bu-hería.
ART-2-elder 3PL-see ART-14-child
‘The elders look after the children.’
b. O-bu-hería ba-bóina a-ba-kairè.
ART-14-child 3PL-see ART-2-elder
‘The children look at/watch the elders.’
In another point of interest when it comes to the use of visual verbs

within the domain of social interaction expressing the sense ‘take care,’
Schneider-Blum & Dimmendaal (2013) present the use of the verb náh
‘see’ in Tima (Kordofanian, Sudan): while the verb náh ‘see’ has more
abstract extensions including a sense of taking care as shown in (4b), not
all expressions can be addressed to blind people. The authors mention that
the expression illustrated in (4a) was uttered, when a child did not pay
attention to a coming car and it can be only addressed to people who are
able to see, while the expression in (4b) may be used even when
addressing a blind person (ibid.). In addition, the visual verb náh ‘see’ in
Tima can be used in a more abstract way at the cognitive level, as
exemplified in (4c), expressing the meaning of personal assumption and/or
impression (ibid: 237).
Tima (Schneider-Blum & Dimmendaal 2013: 236-237)6

(4) a. 1áh-àk-á-tán
see-AP-COMPL
‘watch out/pay attention/take care.’
b. 1áh kì-dȑk.
see SG-neck
‘take care (lit: watch the neck).’
c. c-íbóónína cȑ-nah kì-dȑk cȑ-yȑ
SG-girl IMPFV-see SG-neck IPFV-REP
n-ké-màl-nùn
INS-MOD.SG-beautiful-3SG.LOG
‘The girl considers (lit: sees) herself to be beautiful (but in
fact she is not).’
The example (4c) illustrates that the use of the visual verb náh ‘see’ in
Tima accompanied by the noun meaning kì-dȑk ‘neck’ can be used at a
6
more abstract cognitive level: in this case the semantic weight lies on the
personal impression of the entity being the logical subject of the visual
verb (in this case, the girl), which may be false.
In African languages and beyond the visual verbs may be used with a
meaning corresponding to the English terms ‘search’ and/or ‘find’. In
various African languages the visual verbs develop more abstract
cognitive meanings through this semantic path. Yvonne Treis describes
the polysemy of the visual verb xuud- ‘see/look at’ in Kambaata (Cushitic,
Ethiopia) and notes that this visual verb is often with the meaning
corresponding to the English terms ‘check’ or ‘examine’. In this use the
visual verb expresses the implication that the kind of knowledge or
evidence acquired is a result of a controlling agent, irrespective of the
sensory organs involved in this process (Treis 2010: 328). In this use the
visual verb occurs with active morphology and the experiencer is realised
as the syntactical subject. While the semantic path of the visual verb
huud- ‘see’ in Kambaata leads to the semantic extensions ‘see/look
atĺcheck/examineĺconsider/take into account’, the author notes that the
use of the visual verb for the expression of cognitive meanings like
‘understand/realise/know’ is uncommon in Kambaata. In the few examples
attested in her corpus, where the visual verb corresponds to the
interpretation of ‘realize’, the experience of the perception verb is not
encoded as the subject on the level of syntax: in these cases the visual verb
is morphologically marked with a passive marker and with a bound
pronoun, which marks the dative object referring to the experience:
Kambaata (Treis 2010: 329)

(5) Hujantoomí-r-u kaa´ll-áno-agg-a
work.1PL.OBJ.REL-M.NOM help-3M.ART.REL-M.COP-M.PRED
xuud-ám-ano-ée-hu
see-PASS-3M.ART-1SG.OBJ-M.NOM
‘I realise (lit: it is seen to/for me) that what we did is useful.’
What appears to be common in the way visual verbs are used to

express cognitive meanings in African languages is that the knowledge
associated with vision is of a specific kind linked to visual perception,
which is evaluated lower than other types of knowledge. This type of
knowledge can be expressed through other lexemes and/or through other
verbs of perception. Brenzinger & Fehn (2013) provide examples in
Khwe-||Ani (Khoisan, Namibia) to illustrate this point, as shown in (6).7
7
The Roman numerals in the gloss denote different verbal classes in Khwe.
94 Chapter Seven
The authors first show the use of a visual verb mࠪǊ ‘see’ accompanied by
a verb mਸ ‘know’, which creates the meaning of ‘realizing/identifying’ as in
(6a). They underline the perceptive character of the visual verb by
contrasting the sole use of the visual verb and the use of the visual verb
accompanied by the cognitive verb within the same utterance. (6b)
illustrates that “visual perception may not necessarily extend to any kind
of knowledge for ordinary Khwe” (Brenzinger & Fehn 2013: 185).
Khwe-||Ani (Brenzinger & Fehn 2013: 184)

(6) a. Tí mࠪǊ-a mਸ jWq Ϭíyo à
1SG see-II know-I-PRS snake OBJ
‘I recognize (identify) that snake.’
b. Tí mࠪǊ-jWq támà ti mࠪǊ-a mਸ jWq YH
1SG see-I.PRS but 1SG see-II know-I-PRS NEG
‘I see, but I do not recognize (I cannot identify).’
The extension of the domain of vision to the domain of cognition in

African languages has often a direct visual aspect and an activity-oriented
character. In Dongolawi the visual verb nal ‘see’ can be used in serial
constructions, in which the visual verb appears always as the last
component. In these constructions the meaning of the verb can be
approximately translated into English with the terms ‘check/find out’.
Vision-based evidentials in African languages

Another link between the domain of vision and the domain of cognition
consists of a sensory kind in evidential systems (Aikhenvald 2004), in
which verbs of perception are used as a source for evidentials. Among the
seven languages in Africa, 8 in which grammaticalized evidentials or
evidential strategies are reported only in two of them, in Lega and !Xun,
evidential markers are based on the domain of vision.
In Lega (Bantu, DR Congo), the evidential marker ampo directly
covers the meaning of sight and it is used to express a first-hand
witnessing, even if the source of perception in a given situation is a sound,
as illustrated in (7a). In addition, strong indirect evidence can be expressed
by the same marker as shown in (7b).
8
These are: Lega, Lushese (Bantu), Luwo, Shilluk, Tima (Nilo-Saharan), Sissala
(Niger-Congo), and !Xun (Khoisan).
Concerning the first-hand evidential mèká in !Xun, König explains that

“the first-hand evidential underlines the truth value of the statement,
paraphrasable as ‘Since I have witnessed it with my own eyes there is no
doubt that the following is true’” (König 2013: 75). This marker is often
used to highlight the directness of the source of information.
König points out that in Africa grammaticalized evidentiality seems to
be nearly absent. She underlines that it remains unclear if “the rare
occurrence of evidentials in Africa is due to lack of information or to the
fact that languages indeed have no grammaticalized evidential markers”
(König 2013: 72).
Lega (Bottne 1997: 517-518)

(7) a. Ampó ũ̗kurúrá mompongũ.
EVID pound.3SG.PRS rice
‘She’s assuredly pounding rice (I can hear it).’
b. Ampó MŢkũ̗ ũ̗bilindũ bón
EVID Moke forget.3SG.ART that
Amísi ũ̗ndil ko Pángè.
Amisi go.3SG.PST to Pangi
‘(It is evident, as his odd behaviour indicates, that) Moke
forgot that Amisi went to Pangi.’
Secret visions
While visual verbs in African languages seem to have less extensions to
other domains in comparison with other verbs of perception, the notion of
the invisible as conceived and evaluated in African cultures emerges as a
salient category of perception, which motivates language (cf. Chibaka
2010 and Atindogbe 2010). Anthropological and ethnological studies of
African cultures and societies stress that the concept of seeing the invisible
is often associated with knowledge, wisdom and/or with the ability to
communicate with metaphysical entities. The context of the invisible
vision includes dreams, oramata, oracles and the visit of or possession by
invisible power forces, often called spirits. In Western Uganda spirits
emerge as personalized forces, who are invisible, but able to materialize
and embody themselves (Behrend 2011: 169). Further in the context of
medicine the notion of vision plays a role: often the power of healing is
associated with the knowledge of the abstract powers of life and death. As
reported for numerous cultures in Africa, the power of healing is
associated with the capacity to know about the invisible world. Among
other cultural practices visions and dreams open the door for this mostly
96 Chapter Seven
secret knowledge. The presence of the invisible metaphysical entities is

often announced by smoke, odours or an unnatural shadow (a person who
has no shadow or a shadow which appears although nobody is physically
there). The elusive nature of smoke, odours and shadows underline that
metaphysical powers can´t be visible as any entity in the physical world,
but still their presence is accompanied by signs, which the ones who know
may recognize.
Among the Bashese, the speakers of Lushese in Eastern Uganda, the
invisible world is evaluated as more important than the visible world.
While the holistic and the olfactory verbs of perception are used in
Lushese to express a variety of meanings and have numerous extensions to
the domains of emotion, cognition and social interaction, the visual verb
boina ‘see’ morphologically marked with a reflexive subject-prefix (ee-
boina) denotes the meaning of forgetting. Further, the reflexive form of
the visual verb which expresses forgetting as well reflects the prominence
of social interaction in the local conceptualization: knowing is conceived
as an intersubjective matter, hence the image of someone looking at
his/herself per analogy serves the meaning of forgetting and/ or not
knowing anymore. The following proverb illustrates both the reflexive
forms of the visual verb, which produces the meaning of forgetting as well
as the importance of social interaction among the Bashese.
Lushese (Thanasoula 2016: 290)

(8) A-hée-boina o-mu-bíri, a-tá-boina mu-inè:
3SG-REF-see ART-3-body 3SG-NEG-see 1-friend
a-héeta Málumbè.
3SG-call Malumbe
‘The one who takes care only of his own and neglects others is
going to die.’ (lit. ‘The one who looks at the own body and does
not look at the friend: He is calling Malumbe.’)
Note that Malumbe is the spirit of death, an entity that no clever person
would ever call. Precisely the evocation of the spirit that brings death
underlines the idiotic nature of a selfish person, as expressed in the
proverb. Not only intelligence in general but particular domains of
cognition like knowing, remembering, thinking and understanding are
conceived both as mental and social capacities that link the individual
person with the community and culture, outside of which nobody is able to
survive.
The evil eye

Coming to the cultural scripting of the human body, little is known about
the role of the eye in African languages. Body parts are often used as
sources for grammar, for example, for spatial orientation. Concerning the
eye a common path of grammaticalization associates the eyes, the face
and/or the forehead as sources for the meaning ‘front view’ (for a
summary on the grammaticalization of body parts in African languages
and beyond see Heine & Reh 1984). In this section I would like to treat
another notion concerning the eyes, which is described in numerous
cultures throughout the world and seems to be common among various
African cultures: the concept of the evil eye. The idea of the evil eye roots
itself in body images that associate processes and/or states of the inner self
with body parts and physical abilities of the body, e.g. motion. Especially
in the domain of emotion the linguistic means used to express emotional
events often involve bodily actions, like shivering or various ways of
embodiment (for a detailed analysis on the relations of language, emotion
and embodiment see Zlatev et al 2012).
In the context of the evil eye, the visual organ, which humans are able
to control and move intentionally in different directions, serves as a source
of metaphor expressing personal intention in general. Furthermore, in a
discourse of danger the eye embodies the negative and/or dangerous
intention of somebody, an intention which disturbs the social balance. In
numerous African languages the eyes are used in the context of negative
feelings, especially for jealousy/envy. A summary of these issues is
provided in Ameka (2002). The author gives examples of how certain
body motions, the psychologised eye and red eyes are used in Ewe (Niger-
Congo, Ghana and Togo) to express meanings of jealousy and envy - the
expressions in Ewe involve a verb of motion and the visual organ as in
(9a) or the expression ‘red eye’ as in (9b-c).
As the author points out, the connection between eyes and negative
feelings is common in numerous languages in West Africa. Despite the
similarities concerning the source of metaphors, the processes associated
with the body parts may differ considerably, as Ameka illustrates by
comparing Ewe and Akan (also Niger-Congo), two languages which are
areal and genetic relatives. Further, the cultural interpretation of colours
and the various combinations of colours with body-parts should be
incorporated in future research, since very little is known about this
domain (on colourful expressions in African languages within the domain
of emotion, see Dimmendaal 2002).
98 Chapter Seven
Ewe (Ameka 2002: 31)9

(9) a. va ƾkù
move eye
‘to be jealous of someone.’
b. bia ƾkù
red eye
‘to be jealous about an entity (persons or things).’
c. é-fé ƾkù bia
3SG.POSS eye red
‘(S)he is covetous/down-cast; (s)he is revengeful’.
Conclusions
The lack of information on the language of perception in Africa does not
allow at the moment any generalizations concerning the way perception in
general and vision in particular are expressed in African languages.
Perception is a domain in which specific cultural values as well as social
practices have a considerable impact on the way linguistic means are used
to express this domain. Further, the semantic extensions of words
associated with perception depend on and manifest the various local
cultural interpretations of the environment and the human body. More
research in terms of both quantity and quality is needed in order to
approach the domains of perception in Africa. Besides the value of
linguistic documentation itself, which allows us to understand more about
similarities and differences between African languages, future research on
language and perception in Africa will contribute to tackling the following
challenges: are there any interpretations and cognitive metaphors within
the domain of perception that are shared to such an extent among African
cultures that they could be recognized as pan-African? Are there any
linguistic features concerning semantics or structure that are found only in
Africa?
Further research on the language of perception in Africa will help us to
understand better the way language and perception are linked in general.
This might lead linguistics to more comprehensive universals. On the
other hand understanding perception out of one’s own culture presupposes
abandoning hierarchies of knowledge and accepting concepts that appear
very strange in the first place. Out of a state of aporia the linguistic field of
perception has developed innovative and experimental methods in order to
9
investigate linguistic meaning within the social and cultural context.

Scientific research in Africa still (and unfortunately) preserves some
colonial ideas. Linguistic and ethnological research on the domain of
perception will help scientists to get over this, because it motivates a shift
in the manner of fieldwork. Working on the language of perception means
elaborating more on the local speakers and their ways of experience, to
recognize their opinions on linguistic matters and integrate the folk
explanations in the analysis. The domain of language and perception
requires researchers to apply methodical innovations, which are a
necessary step towards the de-colonization of scientific research in Africa.
References
Aikhenvald, A. Y. & A. Storch 2013. Linguistic Expression of perception
and Cognition. In A. Y. Aikhenvald & A. Storch (Eds.), Perception
and Cognition in Language and Culture, (pp. 1-46). Leiden/Boston:
Brill.
Aikhenvald, A. Y. 2004. Evidentiality. Oxford: Oxford University Press.
Ameka, F. K. 2002. Cultural scripting of body parts for emotions: On
‘jealousy’ and related emotions in Ewe. Pragmatics and cognition, 10,
27-55.
—. 2005. “The woman is seeable” and the woman perceives seeing”:
Undergoer Voice Constructions in Ewe and Likpe. Studies in the
languages of the Volta Basin, 3, 43-62.
Atindogbé, G. G. 2010. Naming the Invisible in Bantu Languages of
Cameroon: On the Semantic Coherence of Nominal Class Systems. In
A. Storch (Ed.), Perception of the Invisible: Religion, Historical
Semantics and the Role of Perceptive Verbs, (pp. 347-372). Cologne:
Köppe.
Behrend, H. 2011. Resurrecting Cannibals:The Catholic Church, Witch-
Hunts and the Production of Pagans in Western Uganda. Suffolk:
James Currey.
Botne, Robert. 1997. Evidentiality and epistemic modality in Lega. Studies
in Language 21/3. 509-532.
Belting, Hans 2008. Florenz und Bagdad. Eine westöstliche Geschichte
des Blicks. Munich: Beck.
Brenzinger, M. & A. Fehn 2013. From Body to Knowledge: Perception
aand Cognition in Khwe-||Ani and Ts’ixa. In A. Y. Aikhenvald & A.
Storch (Eds.), Perception and Cognition in Language and Culture, (pp.
161-192). Leiden/Boston: Brill.
100 Chapter Seven
Chibaka, E. F. 2010. The ‘Invisible’ Perception Verbs Comparison in

Mankon and Meta Succession-Induction Traditional Rites. In A.
Storch (Ed.), Perception of the Invisible: Religion, Historical
Köppe.
Dimmendaal, G. 2002. Colourful psi´s sleep furiously. Depicting
emotional states in some African languages. Pragmatics and Cognition
10(1/2): 57-83.
Evans, N. & D. Wilkins 2000. In the mind´s ear: The semantic extensions
of perception verbs in Australian languages. Language, 76, 546-592.
Fabian, J. 1994. Ethnographic Objectivity revisited: From rigor to Vigor.
In: A. Megill (Ed.), Rethinking Objectivity. Durham/London: Duke
University Press.
Heine, B. & M. Reh 1984. Grammaticalization and Reanalysis in African
Languages. Hamburg: Helmut Buske.
Jakobi, A. & E. E. Guzuuli 2013. Perception Verbs and Their Semanttice
in Dongolawi. In A. Y. Aikhenvald & A. Storch (Eds.), Perception and
Cognition in Language and Culture, (pp. 193-216). Leiden/Boston:
Brill.
Kilgour, Maggie 1990. From Communion to Cannibalism. An Anatomy of
Metaphors of Incorporation. Princeton: Princeton University Press.
König, C. 2013. Source of Information and Unexpected Information
in !Xun-Evidential, Mirative and Counterexpectation. In A. Y.
and Culture, (pp. 69-94). Leiden/Boston: Brill.
Kövecses, Z. 2000. Metaphor and emotion. Cambridge: Cambridge
University
—. 2002. Metaphor: A Practical Introduction. Oxford: Oxford University
Press.
Lakoff, G. & M. Johnson 1980. Metaphors We Live By. Chicago: Chicago
University Press.
Lakoff, G. & M. Turner 1989: More Than Cool Reason: A Field Guide to
Poetic Metaphor. Chicago: Chicago University Press.
Schneider-Blum, G. & G. Dimmendaal 2013. Excite Your Senses: Glances
into the Field of Perception and Cognition in Tima. In A. Y.
and Culture, (pp. 217-250). Leiden/Boston: Brill.
Swartz, Marc J. 1992. On the two bodies in Swahili culture. In J. Maw & J.
Picton (Eds.), Concepts of the Body/Self in Africa, (33-55) Beiträge zur
Afrikanistik, Band 43. Vienna: Institute für Afrikanistik und
Ägyptologie der Universität Wien.
Thanasoula, M. S. 2016. Towards a Grammar of the Senses: Perception in

Lushese, Doctoral Dissertation. Cologne: KUPS.
Treis, Y. 2010. Perception Verbs and Taste Adjectives in Kambaata and
Beyond. In A. Storch (Ed.), Perception of the Invisible: Religion,
Historical Semantics and the Role of Perceptive Verbs, (pp. 313-346).
Cologne: Köppe.
Van Beek, W. E. A. 2010. Eyes on Top? Culture and Weight of the Senses.
In A. Storch (Ed.), Perception of the Invisible: Religion, Historical
Köppe.
Viberg, A. 2001. Verbs of perception. In M. Haspelmath, E. König W.
Oesterreicherv& W. Raible (Eds.), Language typology and language
universals, (pp. 123-162). Berlin: Mouton de Guyter.
Zlatev, J., J. Blomberg & U. Magnusson 2012. Metaphor and Subjective
Experience: a Study of Motion, Emotion Metaphors in English,
Swedish, Bulgarian and Thai. In A. Foolen, J Zlatev, U. Lüdtke & T.
Racine (Eds.), Moving Ourselves, Moving Others: Motion and
Emotion in Consciousness, Intersubjectivity and Language, (pp. 423-
450). Amsterdam: John Benjamins.
CHAPTER EIGHT
ON EXPRESSIONS OF VISION AND OTHER

SENSORY PERCEPTIONS IN SERBIAN
JUNICHI TOYOTA AND BORKO KOVAýEVIû
Abstract. This paper examines how perception, in particular vision, is

expressed in Serbian in comparison with other Indo-European languages.
The linguistic expression of perception can be various, but it is commonly
associated with certain structures, e.g. the middle/reflexive voice or
experiencer in an oblique form. These structures are found with certain
verbs, i.e. they are not productive and lexically determined. However,
Serbian behaves differently and certain grammatical markers can be added
to non-perception verbs with alternation of a subject form, e.g. ja jedem ‘I
eat’ can be turned into a perception verb with the dative subject, jede mi se
‘I want to eat (literary ‘to me it eats itself’).’ Serbian demonstrates a high
sensitivity towards spontaneous perception, and this paper looks into how
this is possible and how other Indo-European languages deal with similar
expressions.
Introduction
Perception is something that we unconsciously do in our daily life, and
whatever language in the world we analyse, there are some expressions
associated with human perception. Studies on perception may be
associated with psychology or cognitive science, and it also plays a crucial
role in cognitive linguistics. For instance, spatial orientation differs greatly
according to each individual’s perspective on a scene. Since space forms a
good base for other areas of our cognition, such as metaphorical extension
from space to time, perception at least indirectly affects various areas of
our language use. However, it is surprising that relatively little attention
has been paid to perception within linguistic studies concerning its
structure and meaning. This paper aims to shed light on the diversity of
linguistic expressions of perception, particularly sight. Perception
naturally includes five basic perceptions, i.e. sight, sound, touch, smell and
104 Chapter Eight
taste. However, it is defined broadly here, and other perceptions of

emotional states such as desire, joy, envy, etc. are also included. Our target
language is a South Slavic language, Serbian. The Slavic languages can
provide rich materials for research concerning perception, but comparison
is also made among other Indo-European languages.
This paper is organised as follows: various expressions of perception in
Serbian are presented first, and this section serves as a base for the rest of
the paper. Following this, internal relations in perception are analysed,
involving various expressions of perception found in other Indo-European
languages. Then structural interpretations of perception are analysed,
focusing on the speaker’s perspective on events and transitivity. This
section explains how to deal with a diversity of expressions found in
Serbian as well as Indo-European languages.
Perception in Serbian
Serbian has various means of expressing perception. What creates
variations in structure is the use of case marking, the grammatical voice
(in particular, the middle voice) and to a lesser extent, tense-aspect.
Furthermore, depending on the structure, a structure allows a speaker to
express his/her own personal interpretation of an event concerning
perception. Let us take a look at the variations one by one. Concerning
verbs of the five basic perceptions, the experiencer is in the nominative
case and used as a subject, and the theme is a direct object, in the
accusative. The examples (1) and (5) exemplify the five basic perceptions.
Sight
(1) Na brdu ja vidim ljude
on hill I.NOM see.PRS.1SG people.ACC.PL
‘I see people on the hill.’
Sound
(2) Na brdu ja þujem ljude
on hill I.NOM hear.PRS.1SG people.ACC.PL
‘I hear people on the hill.’
Touch
(3) Na brdu ja oseüam hladnoüu
on hill I.NOM feel.PRS.1SG coolness.ACC.SG
‘I feel coolness on the hill.’
On Expressions of Vision and Other Sensory Perceptions in Serbian 105
Taste
(4) Na brdu ja okusim voüe
on hill I.NOM taste.PRS.1SG fruit.ACC.SG
‘I taste fruit on the hill.’
Smell
(5) Na brdu ja mirišem cveüe
on hill I.NOM smell.PRS.1SG flower.ACC.PL
‘I smell flowers on the hill.’
In a different structure, the experiencer can be expressed in the dative

case, but there are variations in Serbian. A common expression containing
the dative experiencer appears with the copula verb biti and it denotes
perception, as exemplified in (6). This collocation is very strong and
commonly used (Hammond 2005: 129). In addition to this, there is a
specific use of the dative case which often denotes personal attachment or
adversity, as demonstrated in (7) and (8), respectively. In (7), the pronoun
mi denotes a general possession. This use of the dative form suggests that
a possessed object has some personal attachment or importance. Note also
that there are two types of the dative pronouns, i.e. a clitic form mi and a
stressed form meni, and the form used here is the clitic. Contrary to (7),
the dative form of a pronoun nam ‘to us’ in (8b) suggests adversity, i.e. the
experiencer is annoyed by the event expressed in a clause. A clause
without the clitic does not express such adversity, e.g. (8a). By comparing
these two examples, a sense of perception is expressed solely by the
addition of the dative clitic. These types of the dative case have been given
different terms, e.g. sympathetic dative; ethical dative (Berman 1982);
datif étendu ‘extended dative’ (Leclère 1976, 1978); external possessor
constructions (cf. Vergnaud & Zubizarreta 1992; König & Haspelmath
1997; McWhorter 2002: 224-228). What is important here is that the
dative case is closely related to the expression of perception, and it can be
added to practically any sentence to turn a descriptive clause into a
perception.
(6) Meni je hladno

I.DAT is cold
‘I am cold/I feel cold.’ (lit .‘to me is cold’)
(7) Ja znam gde su mi kljuþevi
I.NOM.SG know where are I.DAT.SG keys
‘I know where my keys are.’ (i.e. ‘where are to me keys’)
106 Chapter Eight
(8) a. Beba plaþe noüu

baby cry.3SG at night
‘The baby cries at night.’
b. Beba nam plaþe noüu
baby us.DAT cry.3PL at night
‘The baby cries at night on us.’ (i.e. ‘to our detriment’)
Apart from the case marking, the reflexive pronoun se is often

associated with expressions of perception. This structure falls into the
domain of the middle voice, which in principle denotes a spontaneous
event. This voice form is highly polysemous in function (cf. Kemmer
1993), and as a sense extension forms a spontaneous reading, a sense of
perception can be expressed in the middle voice. In Serbian, some verbs
referring to perception obligatorily require the reflexive pronoun, e.g.
bojati se ‘be afraid, scared’, plašiti se ‘be afraid, scared’, gaditi se ‘feel
sickened’, etc. These verbal phrases cannot exist on their own, and the
reflexive pronoun has to accompany them to make them fully grammatical.
Hammond (2005: 91-92) lists this type of the reflexive pronoun under
impersonal verbs, claiming that this is closely associated with
impersonalisation. Through this function, the middle voice and the passive
voice shares a functional border (cf. Shibatani 1985, Geniušienơ 1987),
and some may consider the use of reflexive in this sense a type of the
passive voice.
The use of a reflexive pronoun with the verbs of the basic five
perceptions in Serbian adds a sense of potentiality, as demonstrated in (9)
to (13). Notice that the verb has to be in the third person form, and its
reading is also related to impersonalisation. These modal meanings are
common in the middle voice (cf. Kemmer 1993), suggesting the link
between perception and the grammatical voice.
Sight (videti ‘see’)

(9) Na brdu se vide ljudi
on hill REF see.PRS.3PL people.NOM.PL
‘One can see people on the hill/People are visible on the hill.’
Sound (þuti ‘hear’)

(10) Na brdu se þuju ljudi
on hill REF hear.PRS.3PL people.NOM.PL
‘One can hear people on the hill.’
Touch (oseüati ‘feel’)

(11) Na brdu se oseüa hladnoüa
on hill REF feel.PRS.3SG coolness.NOM.SG
‘One can feel coolness on the hill.’
Taste (okusiti ‘taste’)

(12) Na brdu se okusi voüe
on hill REF taste.PRS.3SG fruit.NOM.SG
‘One can taste the fruit on the hill.’
Smell (mirisati ‘smell’)

(13) Na brdu se miriše cveüe
on hill REF smell.PRS.3SG flower.ACC.SG
‘One can smell flowers on the hill.’
In addition to these structures, it is possible to have the dative

experiencer and the reflexive pronoun together in a single clause, as
demonstrated in (14b). The verb itself does not denote perception, as
shown in (14a), but the whole structure in (14b) renders its meaning. Since
there is a reflexive pronoun, the verbal structure is in the third person
singular form. Its meaning fits in the domain of perception, but it also
contains a sense of desire, i.e. a deontic modality. This structure is highly
productive, and a lot of verbs can appear in this construction, e.g. Ja jedem
‘I eat’ - Jede mi se ‘I want to eat’; Ja pijem ‘I drink’ - Pije mi se ‘I want to
drink’; Ja trþim ‘I run’ - Trþi mi se ‘I want to run’; Ja þitam ‘I read’ - þita
mi se ‘I want to read’, etc.
(14) a. Ja spavam
I.NOM sleep.1SG
‘I go to bed.’
b. Spava mi se
sleep.3SG I.DAT REF
‘I want to sleep.’ (lit. ‘to me it sleeps itself’)
However, note that when it comes to the verbs of the five basic
perceptions, there is a restriction based on aspect. As demonstrated in (15)
to (19), the a-examples in the perfective aspect are only possible with a
sense of ability. When it comes to the imperfective aspect, a sense of
desire is expressed.
108 Chapter Eight
Sight
(15) a. Vide mi se ljudi
see.PRFV I.DAT REF people.NOM.PL
‘One can see my people.’
b. Gledaju mi se ljudi
see.IPFV I.DAT REF people.NOM.PL
‘I want to see people.’
Sound
(16) a. ýuju mi se ljudi
hear.PRFV I.DAT REF people.NOM.PL
‘One can hear my people.’
b. Slušaju mi se ljudi
hear.IPFV I.DAT REF people.NOM.PL
‘I want to hear people.’
Touch
(17) a. ?Oseti mi se hladnoüa
feel.PRFV I.DAT REF coolness.NOM.SG
‘One can feel my coolness.’
b. Oseüa mi se hladnoüa
feel.IPFV I.DAT REF coolness.NOM.SG
‘I want to feel coolness.’
Taste
(18) a. Okusi mi se voüe
taste.PRFV I.DAT REF fruit.NOM.SG
‘One can taste my fruit.’
b. Kuša mi se voüe
taste.IPFV I.DAT REF fruit.NOM.SG
‘I want to taste fruit.’
Smell
(19) a. Omiriše mi se cveüe
smell.PRFV I.DAT REF flower.NOM.SG
‘One can smell my flowers.’
b. Miriše mi se cveüe
smell.IPFV I.DAT REF flower.NOM.PL
‘I want to smell flowers.’
Serbian demonstrates different constructions to denote a sense of

perception, which may make a sharp contrast with other Indo-European
languages, such as English. Languages like English have relatively few
options in constructions for expressing perception. See Toyota (2013) for a
discussion of this. Now let us turn to an analysis of the internal structures
of structures presented so far in comparison with other Indo-European
languages.
Perception and grammatical relations

Crosslinguistically, perception is normally expressed based on the
description of a logical sequence of events. In other words, the experiencer
in perception, once considered descriptively, is an unvolitional recipient of
an outer stimulus. This is often realised with the dative or allative case, as
already shown in (6). Defining semantic roles can be problematic in some
cases and because of this, some have claimed covering terms for semantic
terms, only differentiating an instigator of an action and a recipient of
effects from an action, e.g. the actor-undergoer relationship (cf. Van Valin
and La Pola 1997), e.g. the actor covers instigator roles such as agent, and
the undergoer, recipient roles such as patient or experiencer, without
making a finer distinction according to the degree of affectedness.
Comparing perception with non-perception, actor and undergoer are
reversed, as demonstrated in (20) from English for the sake of clarity. In
other words, the actor is the subject entity I in (20a), functioning as a
volitional instigator of action, but it is the direct object him in (20b), an
unvolitional source of perception. Note that I in (20b) is the undergoer, a
mere recipient of a visual input and not physically affected. The
differences are summarised in Table 1. Their surface forms are identical,
but functional roles are opposite.
(20) a. I punched him.

b. I see him.
Table 1. Internal structure concerning perception

(20a) (20b)
Subject is doer/recipient. Doer Recipient
Subject affects object. Affected Not affected
Object affects subject. Not affected Slightly as visual input
Degree of affectedness On object very much Not much
110 Chapter Eight
The lack of a volitional instigator is also related to expressions of time,

i.e. a whole clause tends to be more stative without a volitional actor. This
suggests that the examples in (20) have a sharp aspectual contrast, and
(20a) is dynamic, and (20b), stative. There are various tests to see if a
clause is stative or dynamic. See Toyota (2008: 256-260) for details.
Although perception itself is stative, it can be expressed as dynamic. For
instance, the English example (20b) can be put into the progressive aspect,
e.g. I am seeing him. Stative verbs in English do not allow this, e.g. *I am
existing. Thus, the Serbian examples (1) to (5) are more dynamic than
stative in terms of their structure, while other examples from Serbian
presented so far are stative. This may explain why the examples (15) to
(19) have an aspectual restriction, i.e. the perfective aspect is egressive in
principle, and this cannot accommodate stativity which cannot refer to an
end point of action. The imperfective aspect does not show an endpoint,
and a sense of perception fits better with the b-examples in (15) to (19).
This is also reflected in the applicability of the reflexive pronoun. (9) to
(13) show the reflexive pronoun used with the five basic perception verbs,
and the olfactory verb mirisati ‘smell’ in (13) is the only exceptional case
that does not accommodate the reflexive pronoun in a sense of perception.
It, therefore, suggests that this verb is lexically considered as a dynamic,
non-stative verb, and the nominative subject here is agentive.
Among other Indo-European languages, the structures presented so far
from Serbian can be also found, but with different distributional patterns.
The use of the dative or oblique experiencer, i.e. (6), is commonly found.
One such case is shown in Italian in (21), Lithuanian in (22) and Latvian
in (23). In particular, the Lithuanian and Latvian examples involve verbs
which are normally not associated with perception. These examples with
the dative experiencer suggest that perception in Indo-European languages
is often described based on a logical sequence of events.
Italian
(21) Questo mi piace tanto
this I.OBL please.3SG much
‘I like this very much.’
Lithuanian
(22) Nesi-miegojo ir Jonui
NEG-sleep.PST too Jonas.DAT
‘Jonas too did not feel sleepy.’ (lit. ‘to Jonas did not sleep too.’)
Latvian
(23) Kam niet’
who.DAT itch.3SG
‘Who feels itchy.’ (lit. ‘to whom it itches’)
In addition to the case alternation, middle voice-related constructions

can be found. The Classical Greek example in (24), known as a deponent
verb, has the middle voice suffix -mai, although it lacks its active form. It
is indeed the case that there is a difference in form between the middle
voice marker and the reflexive pronoun, but the examples (9) to (12) from
Serbian and the Classical Greek example both refer to perception through
the domain of the middle voice. It is often the case among the Indo-
European languages that the reflexive pronoun can be used with a certain
set of verbs, forming a kind of idiomatic phrase. Serbian seems to be very
flexible in accepting the reflexive pronoun, to the extent that even
perception verbs themselves can also accept the reflexive, as demonstrated
in (9) and (13). The presence of the reflexive pronoun denotes a sense of
perception in other languages, and in the case of another Slavic language,
Czech, shown in (25b), the addition of the reflexive implies the experience
of pleasure in action, i.e. an experiencer enjoys such an action. Its active
counterpart, (25a), does not denote this sense. Unlike the Serbian
examples, the deontic modality is not implied. This instance is comparable
to the adversative reading denoted by the dative case in (8b) from Serbian.
Classical Greek
(24) hédo-mai ‘I rejoice’
Czech
(25) a. Plavu každý den.
swim.1SG.PRS every day
‘I swim every day.’
b. Plavu si každý den
swim.1SG.PRS REF every day
‘I swim every day (and I enjoy it).’
A number of other Indo-European languages use a dative to refer to

adversity or beneficiary, e.g. (7). The examples in (26) are taken from
German. In this case, the beneficiary can be either a speaker (26b) or
somebody else (26c). German still marginally preserves the case marking
system, but what is interesting here is that some Indo-European languages
have this structure by employing the oblique case on the pronouns,
112 Chapter Eight
although they have completely lost the case marking on common nouns.
The example from Dutch in (27) exemplifies one such instance.
Alternatively, a prepositional phrase can be used to denote the same
meaning, as in the case of Swedish in (28) with a preposition på ‘on’.
Contrary to the case in Dutch, English does not use this structure at all,
although both languages have lost the case marking on common nouns.
German
(26) a. Ich wasche meine Haare
I wash my hair.PL
‘I wash my hair.’
b. Ich wasche mir die Haare
I wash me.DAT the.PL hair.PL
‘I wash my hair.’ (i.e. ‘for my own benefit’)
c. Ich wesche dem kind die Haare
I wash.3SG the.DAT child.DAT the.PL hair.PL
‘I wash the child’s hair.’ (i.e. ‘for the benefit of the child’)
Dutch
(27) Men heft hem zijn arm gebroken
one have.3SG 3SG.OBL his arm broken
‘They broke his arm’ (i.e. ‘to his adversity’)
Swedish
(28) Någon bröt armen på honom
someone break.PST arm.DEF on 3SG.OBL
‘Someone broke his arm.’ (i.e. ‘to his adversity’)
The combination of the dative experiencer and the reflexive pronoun,

as shown in (14b), combines the grammatical voice and the reflexive
pronoun, both of which can denote a sense of perception individually. It
seems rather redundant to have both of them in a single clause, but this
structure is productively applicable to a number of verbs. A possible
interpretation of this structure is as follows: the reflexive pronoun turns a
clause into stative, leading to a spontaneous reading often associated with
the middle voice. Along with this, the use of the dative experiencer creates
a sense of perception, as shown in (6) and (21) to (23). This structure is
common among the Slavic languages, and another example is shown in
(29) from Russian, but it is rarely observed in other languages in the Indo-
European family. This suggests that the Slavic languages show the highest
sensitivity to expressions of perception among the Indo-European

languages.
Russian
(29) Mne horosho spit-sja
I.DAT well sleep-REF
‘I feel sleepy very much/I feel like sleeping very much.’
In addition to these, a different means to express perception can be

found in the Celtic languages. These languages often lack lexical verbs
referring to perception, and a common phrase ‘NP is on experiencer’ is
used. This phrase is based on possession, as demonstrated in (30). The
lexical verb of possession is also missing in Celtic languages and the
periphrastic phrase is used. Notice that the preposition used here is ag ‘at’,
and the possessor is expressed in a locative sense. Perception is also
expressed in a similar fashion, e.g. as shown in (31) and (32), eagla ‘fear’
and áthas ‘joy’ are nouns, which can be comparable to a possessed object
in (30), but the preposition indicating the experiencer is ar ‘on’. This use
of ‘on’ is shown in other Indo-European languages, e.g. Swedish in (30),
but such cases normally refer to an adversative meaning. In the case of the
Celtic languages, however, both beneficiary (i.e. (31)) and adversative
((32)) meanings can be expressed in the same structure.
Irish
(30) Tá leabhar agam
be.PRS book at.me
‘I have a book.’
(31) Tá eagla orm

be.PRS fear on.me
‘I fear.’
(32) Tá áthas orm

be.PRS joy on.me
‘I am glad/delighted.’
The constructions presented so far can be distributed differently

among the Indo-European languages, as shown in Table 2. Others in the
table include prepositional phrases and nominal expressions. In addition,
among the Germanic languages, English differs considerably from others,
having only two features, i.e. lexical and others. Note also that this table
114 Chapter Eight
does not indicate productivity, and the features marked for the Slavic
languages are productive, and there are lexical restrictions in other
languages, e.g. features may exist, but they are only possible with certain
verbs or meanings. Thus, it is obvious that the Slavic languages cover a
wide range of structures to express perception, and they are productive.
Table 2. Distribution of structures among the Indo-European languages

Lexical REF DAT/OBL REF+DAT Others
Celtic ¥ ¥
Germanic* ¥ ¥ ¥ ¥
Indo-Aryan ¥ ¥
Romance ¥ ¥ ¥
Slavic ¥ ¥ ¥ ¥
Notes: * English is an exception, and it only has lexical and others (prepositional phrase).
Structural interpretation
Slavic languages in general show sensitivity to a description of situations
in real life, and the speaker’s visibility and direct experience play a crucial
role in deciding one linguistic form over another. An extreme case is
Russian, as exemplified in (33) to (35). Russian makes a sharp contrast
between reality and non-reality, and the difference is clearly marked in the
linguistic form. (33) is a case of possession, but (33a) is a possession of an
abstract concept which does not exist in this world physically, and (33b)
involves a concrete noun, which is realised as a real-world object. (34) is a
case for existence of an object, and the non-existence of an object denoted
by negation is shown by using the genitive case as in (34b). (35) also
involves the alternation of the case, and this is a reflection of a first-hand
experience of an event by a speaker, and if he/she experienced it directly,
the nominative case is used as in (35a), but if not, the instrumental case is
used, as shown in (35b). This type of sensitivity to reality in surroundings,
known as reality orientation (Dust-Andersen 2011), is also reflected in the
expression of perception. Although there is a physiological explanation,
perception primarily deals with the inner system of the human cognitive
system, making a sharp contrast against the outer world. Thus, the
diversity of expressions for perception can be considered as a reflection of
the distinction between the real and unreal world. This is not restricted to
Russian but is applicable to other Slavic languages, as shown in Table 2.
The diversity of structures also allows speakers to deal with various
situations, reflecting a speaker’s attitude or perspective on an event. Those
languages that are relatively poor in structures, such as English, do not
distinguish reality from non-reality, and they do not have an option to
reflect a speaker’s attitudes, and perception is highly conventionalised as a

mere semiotic sign.
Russian
(33) a. Ya imeju mnenie
I have.PRS opinion.ACC
‘I have an opinion.’ (Abstract noun)
b. U menja jest’ kniga
with I.ACC.SG exist book.NOM.SG
‘I have a book.’ (concrete noun)
(34) a. Byla kniga

be.PST.PRT.F book.NOM.SG
‘There is a book.’
b. Ni bylo knigi
NEG be.PST.PRT.N book.GEN.SG
‘There is not a book.’
(35) a. Dozhd’ smyl pyl’

rain.NOM wash.down.PST,PRT.M dust.ACC
‘The rain washed down the dust.’ (first-hand experience)
b. Doshdyom smylo ply’
rain.INST wash.down.PST.PRT.N dust.ACC
‘The rain washed down the dust.’ (second-hand experience)
When perception is viewed from a crosslingiustic perspective, a

problem lies in a discrepancy between structure and meaning. Its structure
has two arguments, e.g. a person who perceives a sensation and an outer
stimulus. The presence of two arguments indicates a typical transitive
clause based on structure. However, when internal semantic features are
considered, transitivity concerning perception is not obvious, i.e. causation
is not so obviously observable. Perhaps due to structure-biased earlier
research trends, overall syntactic features sometimes fool internal semantic
structures. A seminal paper by Hopper and Thompson (1980: 253-254) has
exceptionally noticed this point, claiming that some instances do not meet
our expectation of transitivity mainly based on the English grammar. Thus,
as Hopper and Thompson (1980) or Taylor (2003: 222-246) argue, a
conventional distinction between transitive and intransitive verbs found in
English may not hold true in other languages based on semantic criteria.
According to their definition based on several parameters, He went away
can be more transitive than I like cakes. As demonstrated in Table 3, the
features from a to j have both high (+) and low (-) values, and by counting
the number of high values, one can determine transitivity in each instance.
116 Chapter Eight
Thus, He went away scores more high-value parameters (seven in total)

than I like cakes does (two in total).
Table 3. Parameters of transitivity for various examples1

He went away. I like cakes. He broke the
window.
a. Participants - + +
b. Kinesis + - +
c. Aspect + - +
d. Punctuality + - +
e. Volitionality + - +
f. Affirmative + + +
g. Mode + + +
h. Agency + - +
i. Affected of object - - +
j. Individuation of object - - +
Keys: + = parameter high; - = parameter low
Facing this structure-meaning discrepancy, Toyota (2009) usefully

makes a distinction between syntactic and semantic transitivity. The
former corresponds to the case in English, where the number of arguments
in each clause is the deciding factor for transitivity, and the latter to the
case where transitivity can be measured in terms of gradience, as shown in
Table 3. Judging from a semantic perspective, transitivity involved in
perception is highly intransitive, as I like cakes in Table 3 shows.
Grammatically, how an experiencer is expressed is a way to indicate
whether a perception is treated as a non-agentive mental or emotional state
(with a dative experiencer) or an agentive action (with a nominative agent).
By changing the case marking on the experiencer, or altering between the
active and the middle voice, each sentence can shows a gradient nature of
transitivity. From an Anglocentric perspective, syntactic transitivity
corresponds to a conventional distinction in linguistic analysis so far, but
semantic transitivity is normally used as a basic operational system in
many languages, as extensively discussed in Toyota (2009). The untidy
1
The features in the table are explained as follows: a. Participants, two or more
participants, agent & object or one participant; b. Kinesis, action or non-action; c.
Aspect, telic (with endpoint) or atelic (without endpoint); d. Punctuality, punctual
or non-punctual; e. Volitionality, volitional or non-volitional; f. Affirmative,
affirmative or negative; g. Mode, realis (about reality) or irrealis (non-reality); h.
Agency, agent high in potency or agent low in potency; i. Affectedness of object,
object totally affected or object not affected; j. Individuation of object, object
highly individuated or object non-individuated.
distinction found in semantic transitivity is the result of historical

development, and between syntactic and semantic transitivity the semantic
one first emerged, expressing the energy transfer in a locational sense with
a metaphorical extension of space. The earlier accusative marking, for
instance, is believed to have been derived from the Proto-Indo-European
allative marker (Martinet 1962: 153; Kuryáowicz 1964: 181). The
syntactic transitivity is a result of the conventionalisation of earlier
semantic-based structures.
English is an exception among the Indo-European languages, since it
treats every verb, whether they refer to perfection or not, identically, and
all phrases referring to perception are grammatically treated as action. This
claim is further supported by the use of the passive voice. As shown in
Toyota (2009), languages with syntactic transitivity allow passivisation of
perception verbs. The passivisation in principle requires a highly transitive
active clause, and the presence of the direct object is a sign of transitivity
with syntactic transitivity and therefore, perception verbs can be
passivized, e.g. (36b). On the contrary, and as long as a clause is
semantically transitive, it is possible to passivise a clause even when a
direct object is not present based on semantic transitivity, as exemplified
in (37) from Dutch. Various conditions to use the passive voice are good
indicators to see what type of transitivity is used in each language. Those
languages with semantic transitivity often use the middle voice to refer to
perception, e.g. the Classical Greek example in (24). This is because the
middle voice was initially used to refer to spontaneous events, and thus it
is low in transitivity.
English
(36) a. Many people liked that film.
b. That film was liked by many people.
Dutch
(37) a. De jongens fluiten
the boys whistle
‘The boys whistle.’
b. Er wordt door de jongens gefloten
it become through the boys whistle.PST.PRT
‘There is whistling by the boys.’
Having claimed this, Slavic languages do not have the full-fledged

passive voice (Toyota and Mustafoviü 2006), and the passive voice cannot
be used to determine what kind of transitivity is used. However, once
118 Chapter Eight
transitivity is semantically measured as in Table 3, it is apparent that

various structures for perception in Serbian are mainly semantically
intransitive, as demonstrated in Table 4. As the number of transitive
features shows, the structure with the nominative experiencer is more
transitive (with six to eight high values), and other structures are
intransitive (with two to three high values). The aspect in Slavic languages
can be either perfective or imperfective, and it is obligatorily marked on a
verb. Thus, the aspect in the table is marked as ‘both possible.’
Table 4. Parameters of transitivity for various examples in Serbian

NOM EXP, DAT EXP, e.g. REF, e.g. (9), DAT EXP +
e.g. (1), (2), (6), etc. (10), etc. REF, e.g. (14b)
etc..
a. Participants + - - -
b. Kinesis + - - -
c. Aspect ± ± ± ±
d. Punctuality ± - - -
e. Volitionality + - - -
f. Affirmative + + + +
g. Mode + + + +
h. Agency + - - -
i. Affected of - - - -
object
j. Individuation - - - -
of object
Keys: + = high parameter; - = low parameter; ± = both possible
The lack of much previous research on perception is, perhaps in part,

due to Anglocentricism in linguistic studies in the past several decades. In
English, a superficial linguistic expression of perception is identical to
other expressions, and the presence or the absence of the direct object is
the only concern grammatically, and this view has been applied to
analyses of other languages. Thus, it appears that there is no need to pay
special attention to perception. Interestingly, however, English can provide
a typologically common type concerning case marking once one goes back
in time, as exemplified in (38) from Old English. In this example, the verb
can be considered as a perception concerning pleasure/displeasure, and
there is no nominative subject, and the arguments involved are in the
genitive (cause) and dative (experiencer). These verbs in Old and Middle
English are known as impersonal verbs. There were numerous verbs
considered as impersonal verbs in earlier English (see Pocheptsov 1997
for a comprehensive list), and they can be classified into three major
syntactic types, as listed in Table 5. Type i, i.e. (38), has disappeared now,
but both type ii and iii exist in Present-day English. However, in terms of
their frequency, Toyota (2013) shows that the type ii construction is rather
infrequent, and type iii, where a person/experiencer is expressed in the
nominative case, is more frequent. He argues that this is most clearly
shown in a structure with a so-called adjectival passive, e.g. (39). As he
argues, this is partly due to an anthropocentric tendency in human
cognition, but English is one extreme case where a human entity, although
its functional role is experiencer, has to be expressed in the nominative
case as an actor. Historical changes in structures for perception, i.e. the
disappearance of a structure in (38), are rather rare typologically, and this
shows a case of grammatical peculiarity in English from a typological
perspective (cf. Toyota 2009 for other examples).
Old English
(38) Mæg þæs þonne ofþyncan ðeodne
may that.GEN then displease.INF lord.DAT
Heaðobeardna
Heathobards.GEN
‘The lord of the Heathobards may not feel pleasure with that.’
(Beo 2032)
Table 5. Types of impersonal verbs in OE and ME

Type i. DAT/ACC - GEN/ACC/PP - V - Clause
‘experiencer’ ‘cause’ ‘neutral’ ‘cause’
Type ii. NOM - DAT/ACC - V - Clause
‘cause’ ‘experiencer’ ‘causative’ ‘cause’
Type iii. NOM - GEN/PP - V - Clause
‘experiencer’ ‘cause’ ‘receptive’ ‘cause’
English
(39) a. I was surprised at the noise.
b. He is interested in linguistics.
Nevertheless, an analysis of transitivity seems to be a good indicator

for how perception is realised in a linguistic form, i.e. low transitivity
seems to allow various structures to denote perception, while high
transitivity often sets a limit on diversity in a structure. Also interestingly,
those languages with semantic transitivity allow speakers to express their
attitudes towards an event. Perception has not been previously associated
with transitivity, and this line of research can be applied to different
languages in the world.
120 Chapter Eight
Conclusions
This paper has examined various constructions denoting perception. The
target language is Serbian, which, along with other Slavic languages,
provides us with a rich recourse for analysis in comparison with mother
Indo-European languages. Perception among the Indo-European languages
is expressed in two distinctive ways. A common pattern follows a logical
sequence of events, and perception is considered as a non-volitional,
spontaneous event, and the experiencer is grammatically expressed as a
recipient of an outer stimulus. This structure has the dative or oblique
experiencer, and the verb has a middle voice-related structure, often
realised with a reflexive pronoun in Indo-European languages. This means
that perception in this pattern is treated somewhat differently from other
structures, and this pattern is widespread among the Indo-European
languages. Contrary to this pattern, some exceptional cases rely heavily on
the structure, and the experiencer and other volitional agents are
grammatically identical in form, and a specific structure is not provided
for perception. English is a notable exception, i.e. English has
conventionalised expressions concerning perception and pays little
attention to a logical sequence of events involved in perception. The
distribution of the different structures varies among the Indo-European
languages, as shown in Table 2, but Slavic languages have a wide range of
structures which are productive.
Perception verbs have attracted little attention from researchers in
linguistics, and this is perhaps due to Anglocentricism in research. English
has a unique way of expressing perception, and since the analysis of
English has been dominant in research, various interesting points
concerning perception have been overlooked. Crucial features here are
two-fold, i.e. the representation of a speaker’s attitudes and how
transitivity is realised in grammar. Various structures mean that speakers
have options in expression, reflecting a speaker’s attitudes. Concerning
transitivity, most languages in the Indo-European family use semantic
transitivity, and thus, the experiencer is not fixed as a single form. This
results in a diversity of structure. Syntactic transitivity, on the contrary,
does not allow this type of flexibility, and perception does not have a
special grammatical marking. Thus, a closer look at transitivity helps us to
comprehend better how perception is linguistically expressed.
References
Berman, R. 1982. Dative marking of the affectee role: data from modern
Hebrew. Hebrew Annual Review, 6, 35-59.
Durst-Andersen, P. 2011. Linguistic Supertypes: A cognitive-semiotic
theory of human communication. Berlin: Mouton de Gruyter.
Geniušienơ, E. 1987. The Typology of Reflexives. Berlin: Mouton de
Gruyter.
Hammond, L. 2005. Serbian: An essential grammar. London: Routledge.
Heine, B. & T. Kuteva 2005. Language Contact and Grammatical Change.
Hopper, P.J. & S.A. Thompson 1980. Transitivity in grammar and
discourse. Language, 56, 251-339.
Kemmer, S. 1993. The Middle Voice. Amsterdam: John Benjamins.
König, E. & M. Haspelmath 1998. Les constructions à possesseur externe
dans les langues d’Europe. In J. Feuillet (Ed.), Actance et Valence dans
les Langues d’Europe, (pp. 525-606). Berlin: Moutin de Gruyter.
Kuryáowicz, J. 1964. The Inflectional Categories of Indo-European.
Heidelberg: Winter.
Leclère, Ch. 1976. Datif syntaxiques et datif éthique. In C.J. Chevalier &
M. Gross (Eds.), Méthodes en Grammaire Française, (pp. 73-96).
Paris: Klincksieck.
Leclère, Ch. 1978. Sur une classe de verbes datifs. Langue Française, 39,
66-75.
Martinet, A. 1962. A Functional View of Language. Oxford: Clarendon
Press.
McWhorter, J. 2002. What happened to English? Diachronica, 19, 217-
272.
Pocheptsov, G.G. 1997. Quasi-impersonal verbs in Old and Middle
English. In J. Fisiak (Ed.), Studies in Middle English Linguistics, (pp.
469-488), Berlin: Mouton de Gruyter.
Shibatani, M. 1985. Passive and related constructions: a prototype analysis.
Language, 61, 821-848.
Taylor, J. 2003. Linguistic Categorization. Oxford: Oxford University
Press.
Toyota, J. 2008. Diachronic changes in the English passive. Basingstoke:
Palgrave.
—. 2009. Passive as an indicator of alignment change: in terms of
transitivity. Groninger Arbeiten zur germanistischen Linguistik, spezial
issue The passive in Germanic languages, 41-52.
122 Chapter Eight
—. 2013. On so-called adjective passive in English. Belgrade English

Language and Literature Studies, 5, 11-48.
Toyota, J. & M. Mustafoviü 2006. Grammatical Voice and Tense-Aspect
in Slavic. In W. Abraham & L. Leisiö (Eds.), Passivization and
Typology, (pp. 191-212). Amsterdam: John Benjamins.
Valin, R.D., jr., Van & R. J. La Polla 1997. Syntax: Structure, Meaning
and Function. Cambridge: Cambridge University Press.
Vergnaud, J.R & M.L. Zubizarreta 1992. The definite determiner and
inalienable constructions in French and Italian. Linguistic Inquiry, 23,
593-652.
CONTRIBUTORS
Borko Kovaþeviü is an assistant professor at University of Belgrade

(Serbia), at the Department of general linguistics. His research interests
cover a wide range of linguistic disciplines, including typology,
morphology, syntax and sociolinguistics, and he has published numerous
articles in these fields. He is also one of the editors for a journal Anali
Filoloskog Fakulteta.
(borkoko@fil.bg.ac.rs)
Ian Richards is an associate professor at Osaka City University (Japan),

in the department of English and American language and literature. His
research interests lie in the field of Post-Colonial Studies. He has
published several books on New Zealand literature, notably To Bed at
Noon: The Life and Art of Maurice Duggan (1997, Auckland University
Press), and Dark Sneaks In: Essays on the Short Fiction of Janet Frame
(2004, Lonely Arts Publishing).
(richard@lit.osaka-cu.ac.jp)
Marina Shchepetunina is a lecturer at Osaka University (Japan). She

works within a field of cultural anthropology, and her research interests
are mainly concerned with mythology and gender studies. She has worked
on ancient Japanese mythology, focusing on Kojiki and Nihon shoki, and
carried out gender studies on divine beings. She has written several journal
articles on topics such as image of Japan as a miko country, shamanism in
Ancient Japanese Myths, and myths and Legends about Ubuya.
(shchepetuninamarina@hotmail.com)
Marilena Stuwe-Thanasoula is a lecturer at the Institute for African

Studies and Egyptology, University of Cologne, (Germany). Her teaching
and research areas include areal typology, cognitive and anthropological
linguistics as well as language policy, language and theatre and language
and gender in Africa. She has published numerous articles on the topics
language and perception, emotion and cognition focussing on African
languages.
(marilena_thanassoula@web.de)
124 Contributors
Junichi Toyota is an associate professor at Osaka City University (Japan),

in the department of English and American language and literature. His
research interests cover a wide range of disciplines, including linguistics,
anthropology and cognitive science. He has published numerous articles
on historical linguistics and cognitive linguistics, and has published
several monographs: Diachronic Change in the English Passive (2008,
Palgrave); Kaleidoscopic Grammar (2009, Cambridge Scholars Press);
The Grammatical Voice in Japanese (2011, Cambridge Scholars Press),
and Sense of emptiness: an interdisciplinary approach (2012, Cambridge
Scholars Press)
(toyotaj@lit.osaka-cu.ac.jp)
Daria Vinogradova is a Ph.D. student at Osaka University (Japan). Her

research interests lie in the fields of linguistics and calligraphy. She has
written several journal articles on ancient Chinese characters and modern
pictograms.
(dariavinogradova@hotmail.com)
INDEX
adjectival passive 119 hearer orientation 46-47, 49-50, 53-

anaphora 58-59 54
Anglocentricism 118, 120 Indo-European languages 48, 59,
areal diffusion 65 103-104, 109-111, 113-114, 117,
areal feature 42-43, 50-51, 65 120
augumentative 62 information structure 39, 59, 65
birth 25-28, 30-35 Iraqw 58-60
bronze script 79 Kojiki 25, 27-28, 30, 34
Chinese characters 69, 71, 75 linguistic orientation 39-40, 45, 53-
ancient Chinese characters 69-71, 54
73, 75-77 Manhire, Bill 11, 21-22
classifier language 51-52 mental vision 45
cognitive frame 40, 43, 45, 52 metaphor 2, 76-77, 85-86, 90, 97-98
counter symbolism 57, 62, 64-65 middle voice 104, 106, 111-112,
Cushitic languages 57-61, 65 116-117, 120
creative power 25-28 moon 25, 28, 35
death 12-13, 21, 30-33, 95-96 mythology 25, 32
deictic 58-59 New Zealand Literature 12, 21
deity 25-26, 28-31, 33-35 Nihon shoki 25, 33
demonstrative 46-47, 49, 57-66 nominal tense 60-61
far distal 58-59, 63-64 non-classifier language 51
distal 57-59, 62-64 nomadic 64
medial 58-59, 64 nomadic pastoralist 64
proximal proximal 57-89, 62-64 null-subject language 39-42, 45, 50-
digital communication 69, 71 52
diminutive 62 ontological 30-31
discourse 39, 42, 58-59, 61, 65, 85- oracle-bone script 71
86, 97 passive 40, 93, 106, 117
emoji 69, 71 perception 2-7, 12, 64, 75, 83-87,
evidentiality 46, 49, 95 89-90, 93-97, 104-107, 109-119
first-hand 49, 94, 114-115 pictogram 69-70
non-first-hand 49 modern pictogram 71-77
eye 5, 7, 19, 21, 25-26, 28, 34-35, pictorial word 70
47, 65-66, 74-75, 83, 90, 95, 97-98 poetry 13, 15-16, 21
ethnosyntax 64 possession 47, 73, 88-89, 95, 105,
existential 5, 47-48 113-114
experiencer 93, 103-105, 107, 109- reality orientation 45, 47, 49, 53-54,
113, 116, 118-120 114
focus 59, 65
126 Index
reflexive pronoun 106-107, 110-112, speaker orientation 46-47, 49, 52-54

120 subject (grammatical) 40-43, 45, 47-
rejuvenation 31-32, 35 48, 51-54, 91-93, 96, 103-104, 109-
replication 65 110, 118
representation of movement 75 sun 25-28, 35
sense of emptiness 40 taboo 25, 29-35
sight 11, 13, 21, 25, 94, 103-104, taste 2-3, 104-105, 107-108
106, 108 topic 59
size touch 26, 29, 103-104, 107-108
physical size 61, 64-65 transitivity 104, 115-120
visually perceived size 64-65 semantic 116-118, 120
smell 2-4, 46, 103, 105, 107-108, syntactic 116-117, 120
110 Tregear, Edward 11-13, 20
Smithyman, Kendrick 16-19 visual language 78
Somali 57-61, 63-65 visual symbol 71-72, 74, 78
sound 13, 46, 57-58, 61-63, 94, 103- vocal cavity 62-63
104, 106, 108 vowel 57-58, 61-64, 66, 90
sound symbolism 57-58, 61, 63-66 high front vowel 57, 62-63, 65
South Island Myth 11, 14-17, 19-22 high back vowel 57, 62-63, 66
South Slavic language 104 worldview 26, 35, 52
spatial composition 71-74, 77-78

Junichi Toyota, Ian Richards, Borko Kovačević, Marina Shchepetunina - Vision Beyond Visual Perception-Cambridge Scholars Publishing (2017)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Junichi Toyota, Ian Richards, Borko Kovačević, Marina Shchepetunina - Vision Beyond Visual Perception-Cambridge Scholars Publishing (2017)

Uploaded by

Copyright:

Available Formats

Vision beyond Visual

Edited by Junichi Toyota, Ian Richards, Borko Kovačević

This book first published 2017

Cambridge Scholars Publishing

Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK

British Library Cataloguing in Publication Data

Copyright © 2017 by Junichi Toyota, Ian Richards, Borko Kovačević,

ISBN (10): 1-4438-8814-1

Acknowledgements ................................................................................... vii

Chapter One ................................................................................................. 1

Part One: Socio-cultural Studies

Chapter Two .............................................................................................. 11

Chapter Three ............................................................................................ 25

Part Two: Cognitive-Semiotic Studies

Chapter Four .............................................................................................. 39

Chapter Five .............................................................................................. 57

Chapter Six ................................................................................................ 69

Part Three: Anthropological and Linguistic Studies

Chapter Eight ........................................................................................... 103

Contributors ............................................................................................. 123

Index ....................................................................................................... 125

The editors wish to express grateful thanks to Margaret Edgcumbe for

DIFFERENT VIEWS ON VISION

JUNICHI TOYOTA AND IAN RICHARDS

Linguistics, anthropology and perception so far

Figure 1. Hierarchical order of perception

HEARING SMELL -contact

TOUCH TASTE +contact

Figure 2. Semantic extensions in perceptual modalities (Viberg 1984: 147)

metaphorical extension referring to cognition, e.g. ‘I hear your point’

Pitjantjatjara (Australian, Evans and Wilkinson 2000: 563, 564)

Figure 3. Semantic extensions across perceptual modalities in Australian

A similar interesting case related to local religious beliefs is found in

Figure 4. Lussesse (Bantu) perception (Thanassoula 2013: 255)

As shown here, previous research on perception is normally in

multidimensional topic, but purely linguistic analysis dealing with internal

Literary studies and reflection on culture

Significantly, Sartre begins with a visual image in his efforts to break

As shown here, cultural issues will always complicate perception of any

Topics covered in this volume

NEW ZEALAND’S SOUTH ISLAND MYTH:

conventional Christian heaven, Tregear reasons, then the afterlife would

‘tis a songless land

The 1930s and 1940s

Zealand European) responses to the new landscape. 1 The New Zealand

Awareness of what great gloom

Man must lie with the gaunt hills like a lover,

The 1950s and 1960s

urban cognoscenti, is Kendrick Smithyman’s ‘Colville’, which was written

That sort of place where you stop

as their hold on small repute,

It could be plainly simple

A school, a War Memorial

The poem begins in a dismissive tone, describing the town of Colville

It could be plainly simple

poet’s consideration of the bay, ‘wise as contrition, shallow/ as their hold

published in 1966, uses its opening stanzas to describe an impoverished

Beyond the 1960s

New Zealand landscape poetry had moved on to other themes, usually

The land itself is only

New Zealand’s Maori name, Aotearoa, is usually translated as ‘The land

THE POWER OF VISION

Abstract. In a mythological worldview, the act of seeing is attributed to