Lecture 2: Historical Discourse Analysis

Introduction
Years ago, people did not really study how our ways of talking and writing
have changed over time. However, since a big book by van Dijk in 1985,
many researchers have started to explore this area. They have called it
different names like "New Philology" and "historical discourse analysis."
These studies look at how people communicated in the past in different
languages and think about why this kind of study is important. They try to
connect the study of language history with the study of how we use
language to communicate, showing how these two areas can help each
other and maybe even come together.

The development of this field has been influenced by three key factors. The
initial breakthrough came from addressing the issue of data scarcity.
Historically, the challenge in analyzing past dialogues and spoken narratives
was the absence of recorded conversations from ancient times. However,
the introduction of extensive and focused digital collections, which offer
easier access to historical texts resembling spoken language, has
significantly changed the situation. This advancement has enabled
researchers to explore written works from the past that mirror everyday
speech, overcoming a significant hurdle in historical discourse analysis.

The second and third enabling factors involve the expansions of the fields
out of which historical discourse arises. On the one hand, in discourse
analysis, it has increasingly been recognized that the techniques of
discourse analysis can legitimately be applied to written language and
that written texts constitute “communicative acts in their own right”
(Jucker, 2008: 896). On the other hand, historical linguistics has come to
focus more on usage, including every day and ephemeral usage
(əˈfem(ə)rəl) ‫رسي ع الزوال‬, to recognize varieties and genre-specific
conventions, to evoke pragmatic and inferential explanations of change,
and to acknowledge the importance of context.

DR. BUSHRA NIMA RASHID 1

2.1 Discourse Analysis: What It Covers

One of the initial challenges in exploring historical discourse analysis is

defining the scope of discourse analysis itself. Leading works on the subject,
such as those by Stubbs (1983), Brown and Yule (1983), and Schiffrin
(1994), encompass a broad spectrum of concepts. These include
mechanisms of cohesion and coherence, reference mechanisms like
anaphora, organization of information (such as topic/comment and
given/new information, and focus), conversational dynamics like turn-
taking, markers of textual boundaries and emphases, strategies for
maintaining continuity on topics or among participants, the use of
discourse markers, and the structuring of texts into paragraphs or narrative
episodes. Additionally, they examine the processes of inference, suggestion
beyond the literal meaning (implicature), assumed knowledge
(presupposition), conversational maxims, and relevance theory, the
principles guiding cooperative communication, strategies for polite
interaction, and the categorization of different types of speech actions.

The line between discourse analysis and pragmatics is notably blurred.

Both fields, as outlined in works such as Levinson's 1983 study on
pragmatics, delve into overlapping territories; it's not uncommon for one to
be viewed as encompassing the other or vice versa. The consensus suggests
that discourse analysis tends to focus more on the text itself—its structure,
its static nature, and its final form. Pragmatics, in contrast, is oriented
towards the users of language, emphasizing the dynamic process of
creating texts and the intentions behind them. While discourse analysis
often aligns with the study of conversation dynamics, pragmatics is
frequently associated with the theory of speech acts.

However, drawing a clear distinction between them remains challenging.

For instance, discourse markers like "well," "so," or "you know" serve
dual purposes: they organize discourse by signaling changes in topic or
speaker, segmenting narratives, or indicating the importance of certain
DR. BUSHRA NIMA RASHID 2
information, aligning with the goals of discourse analysis. Simultaneously,
these markers fulfill expressive roles, whether subjective (such as
conveying emphasis or the speaker's perspective) or interpersonal (such as
drawing the listener's attention, signaling shared understanding, or
conveying politeness), which are typically the domain of pragmatics as
highlighted by Brinton in 1996. This overlap illustrates the complexity of
separating the two disciplines definitively.

This lecture will not detail every topic within discourse analysis, but
broadly, it is seen as the study of naturally flowing spoken or written
communication, focusing on how sentences connect above the individual
sentence level, according to Stubbs (1983). It looks at the overall structure
and elements that link sentences together rather than just local sentence
features. While not precisely distinguishing between discourse analysis and
pragmatics, this talk will lean more towards examining formal text
structures, like discourse markers, over the conceptual parts of text
meaning, such as presuppositions or conversational rules, or how language
is used. Therefore, it will not cover certain topics in historical pragmatics,
particularly those about changes over time in conversational habits,
politeness expressions, or the organization of speech events.

2. 2 "Structures and Links in Historical Texts: A Glimpse into Historical

Discourse Analysis"
Historical discourse analysis, as an interdisciplinary area, can be explored
from two distinct angles: either by applying a discourse-pragmatic lens to
the study of language history or by adopting a historical viewpoint in the
analysis of discourse and pragmatics, as discussed by Jacobs and Jucker in
1995. This dual approach leads to three potential sub-categories. The initial
category is characterized by employing discourse analysis techniques to
examine the historical development of a language, where the analyst's
primary focus is on a specific historical period of that language.

DR. BUSHRA NIMA RASHID 3

This method could be labeled as authentic historical discourse analysis. Its
advantage lies in its capacity to more effectively elucidate the functions of
numerous aspects found in older texts compared to traditional methods.
However, it is crucial to note that this approach primarily deals with
analyzing the forms of a language at a specific stage in its development,
rendering it essentially synchronous.
The second subcategory involves a synthesis of discourse and diachrony. It
encompasses a study of the changes in discourse forms, functions, and
structures over time. That is, discourse structure is treated on a par with
phonology, morphology, syntax, and semantics as something that changes
and develops over time. This approach may be termed diachronic(ally
oriented) discourse analysis. Within both of these subcategories, two
mappings are possible: (1) from form to function, namely, the explication of
the discourse functions of particular historical forms (over time) and (2)
from function to form, namely the identification of historical forms that are
exponents of particular discourse functions (over time) (cf. Jacobs and
Jucker, 1995: 13–25). These represent semasiological versus onomasiology
approaches ‫( المناهج السيمسية مقابل علم األورام‬Lewis, 2012: 903). Both
subcategories may also involve analysis on either the micro-level (e.g.,
discourse markers, terms of address, performative verbs) or the macro-
level (e.g., discourse genres, speech acts, politeness). Jucker (2008: 898–
902) proposes that there are in fact four possible levels of unit to be
investigated:

(1) “expressions” (interjections, discourse markers, address terms),

(2) “utterances” (speech acts),
(3) “discourses” or genres (language of fiction, of medical handbooks, of
personal letters, dialogues), and
(4) “discourse domains” (“a particular, socially defined domain of
interaction” [Jucker 2008: 901] such as the discourse of science or of mass
media).

DR. BUSHRA NIMA RASHID 4

The third subcategory constitutes an application of discourse analysis to
historical linguistics. It is the study of discourse-pragmatic factors in
language change, or of the discourse motivations behind diachronic
changes on all levels. Such an approach has the advantage of providing
elucidation ‫ توضيح‬of certain changes and a fuller understanding of
diachronic processes of change. It may be termed discourse-oriented
historical linguistics.

2. 3 "Exploring Language Evolution: A Historical Approach to Discourse

Analysis"

Historical stages of a language often contain apparently meaningless words

and particles, empty or repetitive phrases, inexplicable morphological
forms or uses of inflectional forms, seemingly “primitive” stylistic
features, and uncategorizable or odd text types. While traditionally, many
of these features have been viewed as grammatical pleonasms, metrical
expedients, intensifiers or emphatics, colloquialisms, or defects of style
‫ أو عيوب السلوب‬، ‫ العامية‬، ‫ المكثفات أو المؤكدة‬، ‫ وسائل القياس‬، ‫ المجاميع النحوية‬, it
has proved fruitful in recent years to re-examine these features using the
tools of modern discourse analysis. While a major stumbling block to such a
re-examination would appear to be the lack of oral texts from earlier
periods, since discourse analysis has typically been concerned with the oral
medium, with naturally occurring conversations, and oral narratives, this is
no longer considered a serious impediment to historical discourse analysis
‫عائق أمام تحليل الخطاب التاريخ‬. First, it is generally agreed that earlier periods
of most written languages, especially medieval texts in the Indo-European
languages, are products of the transition from an oral to a literate culture
and, though not oral texts, contain an “oral residue” (Ong, 1984), the
linguistic characteristics of an oral culture. For Fleischman, it is precisely
because discourse analysis is concerned with oral texts that it will explain
many of the features of medieval literature: “I am convinced that many of
the disconcerting properties of medieval vernacular texts . . . can find more
satisfying explanations if we first of all acknowledge the extent to which our

DR. BUSHRA NIMA RASHID 5

texts structure information the way a spoken language does, and then
proceed to the linguistic literature that explores the pragmatic underpinning
of parallel phenomena in naturally occurring discourse” (1990: 23). Second,
much can be deduced about the oral form of earlier languages from
“speech-based” genres (Biber and Finegan, 1992) such as court records,
sermons, and dramatic dialogue as well as from more colloquial written
genres such as personal letters. Finally, it has become increasingly common
to apply the techniques of discourse analysis to written texts and to
recognize separate principles of discourse structure in such texts: “written
texts can be analyzed as communicative acts in their own right” ( Jacobs
and Jucker, 1995: 10).

2.4 Discourse Markers

In historical discourse analysis, perhaps the most attention has been paid to
what Longacre terms “mystery particles,” that is, to the “verbal and
nominal affixes and sentential particles [which] continue to defy analysis
even at a relatively advanced stage of research” (1976: 468); in
contemporary discourse analysis, mystery particles are more typically
termed discourse markers (Schiffrin, 1987) or pragmatic markers (Brinton,
1996: 29–30, 40) and include such forms as well, now, so, and y’know in
Modern English. As viewed traditionally, discourse markers are of
indeterminate word class and uncertain meaning. But as Longacre
observes, mystery particles almost inevitably “have a function which
relates to a unit larger than the sentence, i.e., to the paragraph and the
discourse” (1976: 468).
It has been convincingly argued that a number of particles can be
understood as functioning as discourse markers with textual and
interpersonal functions; here, space permits only a sampling of articles
discussing particles in the history of the Germanic and Romance languages.
discourse markers tend to take up the initial position in an utterance. This
position is important as it:
- functions as a hint to the status of the discourse marker

DR. BUSHRA NIMA RASHID 6

- serves pragmatic and interactional purposes
- contains the theme of the sentence/utterance (introduces topics, relates
what is being said to the preceding text) (Aijmer, 2002, 29)

Discourse markers can be found not only in the initial position. Depending
on the information structure they can occupy several positions: at the
beginning of an utterance, as “insertions” in the utterance or at the end.
e.g., a) Blanche: Oh, I feel so good after my long, hot bath, I feel so good
and cool and – rested! (TW, 1432)
b) Blanche: Of course, he – he doesn’t know – I mean I haven’t informed
him – of my real age! (TW, 1419)
c) Blanche: I’ve got to write it down – the message, I mean…

Bravo Cladera (2001) analyzed the use of the discourse markers in the
Spanish language dialogue. She found that discourse markers appear in
three positions. They initiate the turn at talk – initial markers (IM), they
mark the central part of the turn – corpus markers (CM), and they mark the
end of the turn – terminal markers (TM). e.g., Natalia: But sure, we want to
do it, aaand … I don’t want to start at the university at once, because I want
to take it easy, for a year at least, aaaaand besides I don’t’ know what to
study. Alcira: And besides, it’s difficult. (translated into English, Bravo
Cladera, 7).

In the example above and besides in Alcira’s turn is an IM, and, because and
‘and’ besides in Natalia’s turn mark the central part of the turn. In the
following example then marks the end of the turn.
e.g., Veronica: and you, what did you do?
Carlos: me, um… nothing. I stayed at home. The telephone was
disconnected, so when my friends called, they could not reach me then.
(translated into English, Bravo Cladera, 10).

The scholar (Bravo Cladera, 2001) has analyzed 367 minutes of

spontaneous conversations and found that discourse markers indicating the
central part of the turn have a major frequency of use and contribute in this

DR. BUSHRA NIMA RASHID 7

way to the progression of the conversation. In the second place, she put
the initial markers, conversational role of which is that of being markers “of
reinforcement of dialogical initiative or reactive connector of two acts or
interventions in the dialogue” (op. cit. /in the cited work, 11). The smallest
amount of discourse markers was used in the final position.
The American linguist Fraser (1993) has also contributed to the analysis of
the position of discourse markers in the English language. He found that
these linguistic elements appear in three positions: each discourse marker
may occur sentence initially, some may take up sentence-medial position
and a few of them may occur in sentence final position.

2.5 Discourse Markers and Proposition

Andersen (2001) in his book “Pragmatic Markers and Sociolinguistic

Variation” questions the assumption that the items that are usually taken
to belong to the category of pragmatic markers (as he calls them) are
external to propositions and do not contribute to truth conditions. He also
shows that some pragmatic markers may affect the truth conditions of
utterances and relates this observation to the grammaticalization and
diachronic development of the forms in question.

The scholar (op.cit./ in the cited work,) argues that pragmatic markers are
interpretable in relation to propositional meaning, which is fundamental to
the interpretation, analysis and understanding of pragmatic markers. It is
not always easy to classify linguistic material as internal or external to
propositions. He argues that some pragmatic markers affect the
propositional meaning of utterances, though not necessarily as conceptual
constituents of propositions but as constraints on their interpretation. The
diachronic development of those items, which become pragmatic markers
is relevant to this issue. The scholar suggests that their problematic status
can be explained with respect to the processes of grammaticalization which
they are involved in.

DR. BUSHRA NIMA RASHID 8

From the point of view of grammaticalization, pragmatic markers are seen
as expressions which, through repetitive use and routinization, have
developed no propositional meanings of a more abstract nature than their
original lexical meanings through processes of conventionalization of
implicatures and increased subjectification.

Diachronic aspects of pragmatic markers are highly relevant. The scholar

(Andersen, 2001) gives pragmatic markers but and like as illustrations to
prove this point. He writes that the present-day conjunction but has
developed from the Old English adverbial and preposition butan with a
spatial meaning of “on the outside, without”. This lexeme is the origin of
both the pragmatic marker but and the preposition but in Modern English,
although only the preposition has a lexical meaning that is similar to the
original spatial meaning. The original lexeme has developed into two
distinct lexemes, one of which encodes a procedure. That is but represents
a case of grammaticalization. As the scholar claims, the diachronic
development of like is analogous. It originates in a preposition with the
meaning “similar to” and has developed into a pragmatic marker. The
lexical predecessor of the pragmatic marker ‘like’ still exists as a
preposition.

‘But’ causes no problems of classification because its grammaticalization

has been completed: it has developed into clear polysemes. As regards
‘like’, the
grammaticalization is still an ongoing process.

Aijmer (2002) also argues that many of the features of discourse markers
(especially their multifunctionality) can be explained as a result of
grammaticalization (either complete or ongoing). By grammaticalization
Aijmer (2002, 16) means the process “whereby lexical items or phrases
come through frequent use in certain highly constrained local contexts to
be reanalyzed as having syntactic and morphological functions, and, once
grammaticalised, continue to develop grammaticalised functions”. The
linguist illustrates the process of grammaticalization by ‘indeed’. This word
DR. BUSHRA NIMA RASHID 9
becomes “integrated into the relatively tight lexical field of epistemic
sentence-adverbs” (op.cit., 17) with the meaning “certainly”. Then, it
acquires a contrastive function, especially after but. As a clause-initial
discourse marker indeed has meanings involving elaboration and
clarification of the discourse intent. The development from adverb to
discourse marker, as the scholar writes, reflects a tendency to use
propositional material for the purposes of creating texts and indicating
attitudes in discourse situations and results in an increase of pragmatic
significance and expressiveness.

Following the tradition of pragmatists such as Austin, Grice, Searle and

Sperber and Wilson, Andersen (2001) maintains that utterances are
accountable in terms of propositions and attitudes towards them.
Generally, problems arise with those markers which have a lexical history.
Those pragmatic markers which have developed from the words with
conceptual meanings, and which have not been fully grammaticalised are
the ones that are likely to be difficult to characterize in terms of
propositionally (e.g., like,
sort of, kind of, you know, you see and especially just).

The use of discourse markers for indicating various types of conversational

moves is very common in dialogue although quite rare in expository text.
While there is a distinction between a conversation and a written text,
there is also a distinction between the domains of functions the markers
have in these two discourses.

It is understandable that a single discourse marker can have more than one
function. Some surveys assign the role of cohesive devices to discourse
markers (Schiffrin, 1987), others focus on their role as speaker attitude
expressives (Andersen, 2001), yet others believe that they are devices for
acknowledging and highlighting the speaker-hearer relationship and
increasing politeness.

1
DR. BUSHRA NIMA RASHID
0
Some scholars tend to classify discourse markers in terms of sharply
defined categories, which is not precise, as we know that the markers are
multifunctional and are difficult to put in a particular category. Andersen
(2001) claims that discourse markers are not only multifunctional in the
sense that they can have different pragmatic functions in different
contexts; they are also multifunctional in the sense that they can take up
multiple functions in one and the same context. So, as any other structural
words (e.g., conjunctions) discourse markers fulfil certain functions in
discourse.

Discourse studies have delved into the role of aspectual forms. Following
grounding principles, Hopper (1979: 219–26) suggests that in Old English
narratives, the foreground typically features perfective aspect verbs
indicating singular, dynamic, punctual, and telic events, while the
background utilizes imperfective aspect verbs denoting states or
durative/iterative/habitual atelic processes. Richardson examines other
aspectual forms in Old English, suggesting that "nonperfective" forms, such
as motion, perception, and ingressive verbs accompanied by infinitives,
signal new episodes, accelerate actions for dramatic effect, and establish
point of view. Additionally, the perfect tense in Middle English serves to
mark narrative progression.
boundaries (1994). I argue that ME inchoative gan ‘began’ serves a
demarcating function and slows the narrative down, while perfective anon
‘at once, immediately’ marks salient action and speeds a narrative up
(Brinton, 1996). Finally, a number of studies have also suggested discourse
functions for EModE ‘do’ as a peak marker, information focuser, or event
foregrounder (Stein, 1985a; Wright, 1989).

Fleischman (1990: 36) concludes that tense-aspect forms serve a variety of

important roles in discourse: they may have textual functions (e.g.,
grounding, creating cohesion, marking boundaries, or modulating pace),

1
DR. BUSHRA NIMA RASHID
1
expressive functions (e.g., expressing evaluation or point of view), and
metalinguistic functions (e.g., signaling text type).
2.5.1 Pronominal Forms

Pronominal forms, because of their anaphoric and referential functions,

play an important role in discourse structuring and hence have also
received the attention of historical discourse analysts. For example, it has
been suggested that the demonstrative pronoun this in ME (as in “this
Pandarus”) functions as a foregrounder (Fludernik ,1995; Sell 1985). Work
on EModE (Early Modern English) has attributed a discourse function to the
variant personal pronominal forms you/thou (see references in Stein,
1985b: 348):

Calvo (1992) argues that in addition to negotiating social identities and

expressing attitudinal features, these forms may denote a change in
conversational topic and mark discourse boundaries; similarly, Hope (1994)
sees these forms as having not only a “macro-pragmatic” function in
encoding the differential status of the interlocutors, but a “micro-
pragmatic” function in expressing emotional attitude. Wales (1995) also
sees a discourse role for the generalizing your (i.e., “not your average
person”) in EModE; in addition to its generic or gnomic meaning, it has
various kinds of expressivity: a deictic, focusing function, a second person
discourse awareness, and a generally dismissive tone.

2.5.2 Fixed Phrases and Clauses

A number of the recognized discourse markers in Modern English consist of

phrases (e.g., after all, all right, and stuff like that) or clauses, sometimes
called “comment clauses” (e.g., I mean, you see, that’s right). Thus, it is
not surprising that fixed expressions in older language, in addition to their
function as oral formulae, are coming to be recognized as discourse markers.
For example, OE ^a gelamp hit ^æt and ME then bifel it that ‘then it
happened that’ can best be understood as a metacommentary marking an
episode boundary and expressing the “subsidiary foreground,” the
1
DR. BUSHRA NIMA RASHID
2
instigating event of an episode. OE hwæt ^a ‘what then’ moves the
narrative forward, expressing the fact that the event which follows can be
inferred from the previous event. In contrast, ME what (ho) makes a claim
on the attention of the interlocutor (Brinton,1996).
Moreover, it is possible to find the origin of modern fixed expressions in
earlier stages of a language. Modern English parentheticals such as I
think/suppose/guess (subjective) or it seems (objective) arise in early ME as
I gesse/trowe/deme or it seemeth; in addition to epistemic and evidential
meaning, they serve purposes of intimacy and “positive” politeness (self-
effacement and deference). Nonfirst person epistemic parentheticals (e.g.,
God knows) also arise in early ME as God woot, trusteth me wel, and serve
as an attempt by the speaker to persuade the hearer of the truth of the
utterance. Likewise, the very common Modern English discourse marker,
you know/y’know, arises in ME as ye knowen, perhaps as a replacement for
OE hwæt (see above) (Brinton, 1996).

2.5.3 Word Order

The relation of word order patterns to discourse factors such as
‘topic/comment, thematization, and focus’ is well known. An account of
such phenomena, which have been widely studied in the word order of
older languages, is beyond the scope of this lecture. However, a somewhat
broader view of discourse factors in the word order of an historical
language is taken by Hopper (1979, 1992), who suggests that word order in
OE can be accounted for by a theory of grounding. He argues that the
foreground is characterized by (S)OV or VS (O) (“verb peripheral”) word
order, while the background is characterized by (S)VO word order. In
respect to verb peripheral order, (S)OV is used internal to episodes with
topical subjects and VS (O) is used at the beginning of minor episodes and
with a change in subject or topic. (S)VO is used for the beginning of main
episodes and for global backgrounding.

2.5.4 Text Types

Finally, it has been suggested that typologies accounting for current texts
and the enumeration of features characteristic of different text types may

1
DR. BUSHRA NIMA RASHID
3
not be adequate for a classification of texts from the past, since
conventions of genre are defined by a variety of factors, including forms of
the language, topic, situation, and medium. Fries asserts, for example, that
“it must not be taken for granted that text-linguistic rules for present-day
English are also valid for the older periods of the language” (1983: 1013).
Questions of differences of textual conventions fall under what Jacobs and
Jucker (1995: 11) call “pragmaphilology,” or “the contextual aspects of
historical texts, including the addressers and addressees, their social and
personal relationship, the physical and social setting of text production
and text reception, and the goal(s) of the text.”

2.6 Discourse-oriented Historical Linguistics

The second approach to historical discourse analysis is one which seeks to

find the origins and/or motivations of diachronic change in discourse. This
approach has been ascendant in recent years. Since it would be impossible
to give a complete picture of the results of this approach, this section can
only hint at areas in which these types of studies have concentrated.

2.6.1 Discourse-driven Change

It has become almost standard practice in linguistic research to consider
discourse pragmatic factors as possible causes, motivations, or essential
aspects of historical change. Two areas of change in which discourse
motivations seem most clearly at work are word order change and
grammaticalization.
2.6.2 Word Order Change
It would seem obvious to conclude that just as there is an essential link
synchronically between word order and discourse, there should be such a
link between word order change and discourse. The work of Faarlund
(1985, 1989) on “pragmatic syntax” is typical of this approach to word
order change. Faarlund argues that “the goal [of pragmatic syntax] is to
account for the choices speakers make between systematically related

1
DR. BUSHRA NIMA RASHID
4
surface structures with equivalent cognitive content” in terms of factors
such as theme, focus, and dominance; in other words, whenever two or
more (synonymous) syntactic forms exist, there are pragmatic reasons for
using one rather than the other. He believes that syntactic change can be
explained in terms of pragmatic syntax, for if a new form appears and
becomes pragmatically more useful, it may lead to syntactic restructuring,
or what Faarlund calls the “grammaticalization of pragmatics” (1985: 366–
8, 386). As an example of such change, he discusses the change from OV to
VO word order in Germanic. The rightward movement of the object should
not be explained as a rare and highly marked afterthought, but by a
universal pragmatic principle of focusing.
2.6.3 Grammaticallization
More recently, it has come to be recognized that discourse factors play a
role in the process of grammaticalization. A widely accepted view of
grammaticalization is that rather than involving semantic “bleaching” (loss
of meaning) or metaphor, as has traditionally been assumed, it involves a
change from conversational to conventional implicature; that is, a
conversational implicature arising in certain local discourse contexts
becomes “semanticized,” or assimilated as part of the conventional
meaning of the grammaticalized word. This type of change has been called
“pragmatic strengthening” or “strengthening of informativeness”
(Traugott and König, 1991; Hopper and Traugott, 1993: 63ff; Traugott,
1995b).
Numerous examples of the role of conversational implicature in
grammaticalization have been adduced by Traugott, primarily from the
history of English. An instance of such a semantic shift is the change from
temporal to causal meaning in the grammaticallization of OE si^^an ‘since’
from adverb to conjunction, from the meaning ‘from the time that’ to the
meaning ‘because’, which results from semanticization of the meaning of
‘cause’ which arises in certain contexts. Working within the same
framework, Carey (1994), considering the early grammaticalization of the

1
DR. BUSHRA NIMA RASHID
5
perfect in OE, sees the shift from stative (adjectival) to perfect (verbal)
meaning, that is, from present state of an object to past process performed
on an object, as the conventionalization of an invited inference.

2.7 Diachronically Oriented Discourse Analysis

The third type of historical discourse analysis is one which examines the
evolution of discourse marking over time, whether focusing on the
development of individual discourse markers or on changes in systems of
discourse marking.

2.7.1 The Origin and Development of Discourse Markers

A number of questions arise in the study of the development of discourse

markers:
1. What is the source of discourse forms? What semantic and syntactic
properties predispose them to express certain discourse notions?
2. What is the course of their semantic and syntactic development? Do they
follow recognized principles of change?
3. How do they fare over time? What changes do they undergo and why? To
what extent are they transient?

Most studies of the evolution of discourse markers have related their

development to the unilinear course of grammaticalization proposed by
Traugott (1982: 257), from propositional/ideational to (textual) to
interpersonal/expressive meaning, following three principles of semantic
change (Traugott and König, 1991: 208–9):
• tendency I: from meanings situated in the external described situation to
meanings situated in the internal (evaluative/perceptual/cognitive)
situation;
• tendency II: from meanings situated in the described external or internal
situation to meanings situated in the textual/metalinguistic situation;
• tendency III: to meaning increasingly situated in the speaker’s subjective
beliefstate/ attitude toward the situation.

1
DR. BUSHRA NIMA RASHID
6
Tendencies I and II are metaphorically driven, while tendency III is
metonymically driven, involving an increase in informativeness or a
conventionalizing of conversational implicature (see above). Tendency III
results in “subjectification,” or “the development of a grammatically
identifiable expression of speaker belief and speaker attitude toward
what is said” (Traugott 1995b: 32).

Traugott provides examples of discourse markers such as 'well, right, and

why' evolving in meaning from propositional to textual to interpersonal
functions. Additionally, she illustrates how phrases like 'of let’s' and 'of let
alone' undergo semantic shifts from imperatives to markers indicating
speaker awareness and epistemic attitudes, respectively. Furthermore,
Schwenter and Traugott highlight the evolution of prepositional phrases
like 'instead/in place/in lieu of' from purely locative expressions to markers
implying (counter) expectation. Similarly, Rickford et al. discuss the
transformation of 'as far as' from denoting distance to functioning as a
topic restrictor and discourse marker over time.
Finell (1989, 1992) observes a similar course of development with well in
English and with topic changers, including introducers (now), closers
(however), and resumers (anyhow).
In general, research has found that in their development, discourse markers
undergo many of the morphosyntactic and semantic changes identified
with the process of grammaticalization, though never, of course, being fully
“grammaticalized” in the sense of being incorporated into a recognized
grammatical paradigm nor generally undergoing phonological reduction or
morphological bonding. They are subject to the following changes, all of
which are thought to be typical of grammaticalization:

1. Decategorialization: loss of the morphological and syntactic

characteristics of their original word class);
2. Change from open to closed class membership (Traugott);

1
DR. BUSHRA NIMA RASHID
7
3. Syntactic fixation: loss of syntactic variability and occupation of a fixed
slot (but see Traugott, 1995b);
4. “divergence” (Hopper, 1991) or “split”: retention of full lexical
characteristics in some contexts alongside grammaticalization in other
contexts; and
5. “layering” (Hopper, 1991): continuation of older, more highly
grammaticalized forms next to newer, less grammaticalized forms.

2.7.2 Changes in Discourse Marking

In addition to the evolution of individual discourse studies, attention has

also been paid to larger changes in patterns of discourse structuring, from
one system of discourse marking to another system. For example, Wårvik
(1990) sees a “typological” shift in the history of English from the explicit
foreground-marking system of OE, centered on the use of ^a “then”, to the
“fuzzy” backgrounding system of Modern English, which depends on the
tense-aspect system (simple vs. expanded tenses) and the syntactic status
of clauses; she relates this shift to a change from oral to literate techniques
of grounding (cf. Aristar and Dry, 1982). In contrast, Taavitsainen (1995)
sees interjections, as they become restricted to the oral context, as losing
the textual functions (e.g., reader involvement, turning point in plot,
vividness of narration, topic shift) that they had in EModE, while continuing
the speaker- and addressee-focusing functions.

Despite the changes in discourse forms over time or their loss, there would
nonetheless seem to be a continuity of pragmatic functions over time,
with the forms expressing discourse functions – forms which seem to be
intrinsically ephemeral (see Stein, 1985a) – continually being replaced; this
process of “renewal” is characteristic of grammaticalization (Hopper,
1991).
2.7.3 Changes in Text Types

1
DR. BUSHRA NIMA RASHID
8
Stein (1985b: 351) argues that the study of text types has always possessed
a historical dimension, but previous research on changes in discourse or
genre has predominantly focused on shifts from oral to written
communication. However, Biber and Finegan (1989, 1992) have taken a
broader approach, analyzing changes across various written and spoken
genres in English. They observe a trend where all genres have shifted
towards more "oral" characteristics, such as increased use of private verbs,
first and second-person pronouns, and contractions, as opposed to more
"literate" features. Atkinson (1992) further supports this trend by analyzing
medical research writing from 1735 to 1985, which shows a progression
towards more informational and less narrative styles. Similarly, Görlach
(1992) examines changes in cookery books from Middle English to the
nineteenth century, noting a transition from oral to written traditions,
development of generic conventions, and introduction of social distinctions
in audience targeting.
Given that the results of genre-specific study and cross-genre studies have
shown opposite directions of change in respect to the oral/written
continuum, it seems clear that this area needs much fuller study.
Moreover, the linguistic features defining “oral” and “written” texts need
to be understood better than they currently are before a diachronic study
of texts can come to any certain conclusions. One might also question
whether the focus on oral and written features, given the uncertainties
surrounding this topic, is the most useful one.

1
DR. BUSHRA NIMA RASHID
9