Biological classification a philosophical

introduction 1st Edition Richard A.

Richards
Visit to download the full and correct content document:
https://textbookfull.com/product/biological-classification-a-philosophical-introduction-1
st-edition-richard-a-richards/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Sport A Biological Philosophical and Cultural

Perspective Jay Schulkin

https://textbookfull.com/product/sport-a-biological-
philosophical-and-cultural-perspective-jay-schulkin/

Rule 1 1st Edition T A Richards Neville Richards

Neville T A

https://textbookfull.com/product/rule-1-1st-edition-t-a-richards-
neville-richards-neville-t-a/

Philosophical Logic: A Contemporary Introduction John

Macfarlane

https://textbookfull.com/product/philosophical-logic-a-
contemporary-introduction-john-macfarlane/

Psychology A Concise Introduction 5th Edition Richard

A. Griggs

https://textbookfull.com/product/psychology-a-concise-
introduction-5th-edition-richard-a-griggs/
Sociology: A Brief Introduction Richard T. Schaefer

https://textbookfull.com/product/sociology-a-brief-introduction-
richard-t-schaefer/

Living Philosophy: A Historical Introduction to

Philosophical Ideas 2nd Edition Lewis Vaughn

https://textbookfull.com/product/living-philosophy-a-historical-
introduction-to-philosophical-ideas-2nd-edition-lewis-vaughn/

Beguiling his Timid Waiter A Paranormal s Love Book 25

1st Edition Charlie Richards [Richards

https://textbookfull.com/product/beguiling-his-timid-waiter-a-
paranormal-s-love-book-25-1st-edition-charlie-richards-richards/

Introduction to a Negative Approach to Argumentation

Towards a New Ethic for Philosophical Debate 1st
Edition Julio Cabrera

https://textbookfull.com/product/introduction-to-a-negative-
approach-to-argumentation-towards-a-new-ethic-for-philosophical-
debate-1st-edition-julio-cabrera/

The Philosophical Imagination Selected Essays 1st

Edition Richard Moran

https://textbookfull.com/product/the-philosophical-imagination-
selected-essays-1st-edition-richard-moran/
Biological Classification

Modern biological classification is based on the system developed by

Linnaeus, and interpreted by Darwin as representing the tree of life. But
despite its widespread acceptance, the evolutionary interpretation has
some problems and limitations. This comprehensive book provides a
single resource for understanding all the main philosophical issues and
controversies about biological classification. It surveys the history of
biological classification from Aristotle to contemporary phylogenetics and
shows how modern biological classification has developed and changed
over time. Readers will also be able to see how biological classification
is in part a consequence of human psychology, language development,
and culture. The book will be valuable for student readers and others
interested in a range of topics in philosophy and biology.

RICHARD A. RICHARDS is a Professor of philosophy at the University

of Alabama. His publications include The Species Problem (Cambridge
University Press, 2010) and many journal articles.
Cambridge Introductions to Philosophy and Biology

General editor
Michael Ruse, Florida State University

Associate editor
Denis Walsh, University of Toronto

Other titles in the series

Derek Turner, Paleontology: A Philosophical Introduction
R. Paul Thompson, Agro-technology: A Philosophical Introduction
Michael Ruse, The Philosophy of Human Evolution
Paul Griffiths and Karola Stotz, Genetics and Philosophy: An Introduction
Biological Classification
A Philosophical Introduction

RICHARD A. RICHARDS
University of Alabama
University Printing House, Cambridge CB2 8BS, United Kingdom

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of

education, learning and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781107687844

© Richard A. Richards 2016

This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.

First published 2016

Printed in the United Kingdom by Clays, St Ives plc

A catalog record for this publication is available from the British Library

Library of Congress Cataloging-in-Publication data

Names: Richards, Richard A., author.
Title: Biological classification: a philosophical introduction / Richard A. Richards.
Description: New York: Cambrdige University Press, 2016. |
Series: Cambridge introductions to philosophy and biology |
Includes bibliographical references and index.
Identifiers: LCCN 201601847 | ISBN 9781107065376 (hardback) |
ISBN 9781107687844 (pbk.)
Subjects: LCSH: Biology – Classification – Philosophy.
Classification: LCC QH83.R483 2016 | DDC 578.01/2–dc23
LC record available at https://lccn.loc.gov/2016018473

ISBN 978-1-107-06537-6 Hardback

ISBN 978-1-107-68784-4 Paperback

Cambridge University Press has no responsibility for the persistence or accuracy of URLs
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
In memory of my father, Richard Richards
Contents

List of Illustrations page viii

Acknowledgments ix

Introduction 1

1 Why Classify? 6

2 The Aristotelian Framework 36

3 The Darwinian Pivot 64

4 Evolutionary Taxonomy and the Cladistic Challenge 101

5 The Tree of Life 146

6 The Species Problem 178

7 The Metaphysics of Biological Classification 209

8 Theory and the World 243

9 The Essential Tension 276

References 287
Index 299

vii
 List of Illustrations

4.1 Genera on a Phylogenetic Tree. (From Mayr 1942, 282) page 106
4.2 Semiarbitrary Paraphyly. (From Simpson 1961, 206) 111
4.3 Paraphyly and Reptiles. (From Mayr and Ashlock 1991, 249) 112
4.4 Flow Chart of Numerical Taxonomy. (From Sokal and
Sneath 1963, xviii) 121
4.5 Hennig’s Monophyly. (From Hennig 1966, 71) 127
4.6 Tokogenetic Relations. (From Hennig 1966, 31) 128
4.7 Parsimony-based Cladogram. (From Eldredge and
Cracraft 1980, 25) 135
4.8 Conflicts in Parsimony. (From Eldredge and Cracraft 1980, 68) 137
4.9 Cladogram and Tree. (From Nelson and Platnick 1981, 144) 141
5.1 Sequencing and an Asymmetric Tree 159
5.2 Deep Coalescence. (Courtesy of Alana Baldwin, Multimedia
Services, University of Alabama) 169

viii
Acknowledgments

Michael Ruse made this book possible. I am indebted to him for his exam-
ple, encouragement, and generous support. Hilary Gaskin and Rosemary
Crawley of Cambridge University Press have been helpful and understand-
ing. I am also indebted to Peter Achinstein. I could not have had a better
mentor. My wife, Rita Snyder, has provided support and companionship in
dance and life. Robert Olin, the Dean of Arts and Sciences at the University
of Alabama, and his Leadership Board have been generous in their support.
Scott Hestevold, my chair, colleague, and friend, has also helped make
this book possible. I have benefited from conversation and correspondence
with David Hull, Michael Ghiselin, Richard Mayden, Mary Winsor, and
Matt Haber, and have learned much from the good work of countless oth-
ers. The views presented here, and any errors, are my own.

ix
 Introduction

This is a book about the philosophical history and foundations of modern

biological classification. How and why did we come to classify individual
organisms into all those Linnaean categories – species, genera, families,
orders, classes, and so on? Here biological classification will be understood
broadly, to refer to the comparison and grouping of organisms, the naming
of these groups, the theoretical basis for grouping, and the philosophical
foundations for systems of grouping. On this broad construal, it encom-
passes what is often referred to as ‘taxonomy’ or ‘systematics.’
But why should we care about biological classification? What scien-
tific purpose does it serve? Shouldn’t we instead be focused on important
biological processes, especially those at the molecular level? This last ques-
tion was asked early in the twentieth century, as the biologist Ernst Mayr
lamented:

The rise of genetics during the first thirty years of this century had a rather
unfortunate effect on the prestige of systematics. The spectacular success
of experimental work in unraveling the principles of inheritance and the
obvious applicability of these results in explaining evolution have tended
to push systematics into the background. There was a tendency among
laboratory workers to think rather contemptuously of the museum man,
who spent his time counting hairs or drawing bristles, and whose final aim
seemed to be merely the correct naming of his specimens. (Mayr 1942, 23)

Today, in the age of genomics, genetics, and epigenetics, we might wonder

if these laboratory critics were right. From what we learn about genomes
and genes, we can come to understand the molecular basis for develop-
ment and how this development is affected by environmental factors.
And from population genetics, we can learn how these molecular factors
change over time, producing evolutionary adaptation and diversification.

1
2 Introduction

Isn’t this the future of modern biology, not the old-fashioned, pedantic
categorization and naming of organisms based on mere similarities and
differences? Perhaps classification is necessary, but it is relatively trivial.
But even on this way of thinking, biological classification is hardly
trivial. Surely those who study genomes need to know what kinds of
organisms are providing the genomes. In the most obvious case, the
human genome project studies just those genomes of members of Homo
sapiens. And those who study the genomes of primates in general are
typically interested in the fact that they are primates and share common
ancestry. Moreover, to even know that we are referring to the same kind
of organisms, we need a stable, precise naming system. For that we need
some kind of careful bookkeeping system that associates a particular
name with a careful and precise description that allows the identification
of individual organisms. To study the genes, genomes, and genotypes of
Panthera tigris, for instance, we need a way to identify an organism as a
tiger – to name it. To study the genes and genomes of bacteria, we need a
way to identify an organism as a bacterium. And to study horizontal gene
transfer in bacteria, we need to distinguish the different kinds of bacte-
ria. In all these cases, we need a system of biological classification that is
precise and unambiguous enough to let us know what kinds of things we
are studying.
But the significance of biological classification goes beyond this simple
bookkeeping function. Those who study biodiversity are often interested
most in a single biological category – species. They may measure biodiver-
sity in terms of species counts, and propose conservation efforts to preserve
species. This requires not just that we identify an organism as a particular
kind of living thing, but a kind of thing at a particular rank or level – the
level of species. As we shall see, the determination of species groupings is
far from trivial. There are multiple, conflicting species concepts that divide
up biodiversity in different ways, and no obvious reason to prefer one con-
cept over another. The species level has another significance in that specia-
tion is taken to be an important evolutionary process. Most basically, it is
the appearance of new kinds of living things. But to know when a new spe-
cies is formed, we need to be able to identify the species of organisms, and
we need to know what makes a grouping of organisms a species grouping.
All this requires some sort of theoretical basis for identifying species and
distinguishing the species level from other classificatory levels.
 Introduction 3

What all this suggests is that the study of many biological processes,
even at the molecular level, relies heavily on biological classification, and
not just in a trivial, bookkeeping way. But modern biological classification
is also practiced within an evolutionary framework and is standardly taken
to represent evolutionary history. Organisms are grouped together on the
basis of common ancestry, and classification therefore represents patterns
of evolutionary diversification. To use a powerful metaphor, it represents
the “tree of life.” Modern systematists don’t simply record the similarities
and differences among organisms, “counting hairs and drawing bristles”
in Mayr’s words quoted above. They are often engaged in the complex and
difficult process of reconstructing the evolutionary past. But this process,
as we shall see, raises many questions and is fraught with many difficul-
ties. What can we legitimately assume about biological processes in our
reconstruction of the tree of life? What is that tree like? And is there even
something plausibly described as a tree?
The bottom line is this: Biological classification informs and relies on
many other areas in biology. Rather than being an isolated and pedantic
activity, it is an activity integrated with, and crucial for the practice of mod-
ern biology. Perhaps we can better understand how and why by looking at
it from a philosophical perspective.
In Chapter 1, we begin by looking at the cultural and psychological
foundations for biological classification. All cultures classify organisms and
in broadly similar ways. But this is not surprising, as cognitive psycholo-
gists and developmental linguists tell us that hierarchical classification is
a natural consequence of our cognitive development and language acqui-
sition. But although it may be natural to classify, not all of our classifica-
tions are “natural,” in that they pick out important, human independent
categories in nature. One way to understand this is through the philosoph-
ical framework that distinguishes natural kinds from conventional and
artificial kinds.
An important part of understanding modern biological classification
is through its history. How did we come to classify organisms this way?
Usually the beginnings of modern biological classification are traced to the
thinking of Aristotle, in his use of a method of logical division and the clas-
sificatory terms ‘eidos’ and ‘genos.’ Aristotle has standardly been construed
here as an “essentialist,” identifying biological taxa with a set of unchang-
ing set of essential generic properties and specific differentia. But as we
4 Introduction

see in Chapter 2, this understanding of Aristotle’s views on classification

is, at best, highly misleading. Nonetheless, the Medieval commentators on
Aristotle, who were more interested in logic, language, and theology than
biology, misunderstood him in similar ways.
In Chapter 3, we continue this historical survey by looking at the begin-
ning of modern biological classification in the medical herbalists of the
Renaissance, and the early naturalists, in particular the Swedish botanist
Linnaeus. In this period we see the development of a fixed hierarchy and
a stable system of naming, along with a turn to theoretical questions.
Linnaeus, and other naturalists, including John Ray, Buffon, Kant, MacLeay,
and Cuvier began to ask what made a classificatory system “natural,” and
what it should represent. These questions did not get satisfactory answers
until Darwin’s evolutionary approach: classification should represent
ancestry and the branches on the evolutionary tree.
Darwin’s evolutionary framework was adopted and developed in the
early twentieth century by Ernst Mayr and G. G. Simpson, who worked to
integrate evolutionary classification with new theoretical developments on
heredity and natural selection. But by the mid-century, this “evolutionary
systematics” found challenges, one from a group of radical empiricists, the
“pheneticists,” who objected to an evolutionary basis of classification on
the grounds that we lack sufficient knowledge of the evolutionary past.
A second challenge came from a group of systematists known as “cla-
dists,” and was evolutionary in that they took classification to represent
the branching of the evolution. But cladists, unlike the evolutionary sys-
tematists, rejected the use of assumptions about evolutionary process to
reconstruct the past.
Classification has increasingly come to be seen as representing the evo-
lutionary tree, as we see in the various tree of life projects. But there are
complications here. First, the Linnaean ranking system is inadequate to
represent the many branches of the evolutionary. Second, there seems to
be extensive reticulation in the tree of life, with hybridization and horizon-
tal gene transfer. Third, as there are multiple inconsistent trees represent-
ing species and genes, it is not obvious that there is a single tree of life. In
Chapter 5, we look at these complications, and how we might best think
of evolutionary trees.
The species rank seems to be unique, as species are often described
as the fundamental units of classification and evolution. But there are
 Introduction 5

multiple, inconsistent ways to conceive species. Chapter 6 looks at this

species problem, and the major positions taken about the nature of spe-
cies. One could be a pluralist about species, recognizing different kinds of
species. But some kinds of pluralisms seem to deny the reality of species.
One possible solution is that different species concepts function in differ-
ent ways. However we think about species, there seem to be metaphysical
implications, as reflected in the two main views: species as sets and species
as individuals. In Chapter 7 we look at how these two ways can be extended
to classification at other levels. How should we think about biological taxa
at the most general, fundamental level?
Chapter 8 will turn to a philosophical topic that has lurked beneath
much of the discussion in the previous chapters: What is the rela-
tion between theory and classification? We will begin by looking at the
Baconian Ideal, where observation precedes both classification and theory.
We will then contrast it with a more theoretical approach, where both
observation and classification are theoretically informed. By considering
these two approaches, we can better understand the role evolutionary
theory has played since Darwin’s interpretation of the Linnaean hierar-
chy. We will finish in Chapter 9 with a look at a fundamental tension
revealed in the previous chapters. Our psychology and language learning
incline us to think about biological classification as timeless natural kinds
with essences. But our best theory tells us instead that biological taxa are
historical entities – branches on an evolutionary tree. These two ways
of thinking about taxa – as timeless kinds and historical entities – seem
inconsistent, and have been implicit in the more than two-thousand-year
history of biological classification. Until human psychology or the evolu-
tionary basis changes, biological classification will be fraught with this
fundamental tension.
 1 Why Classify?

Classification and the Diversity of Life

Nature is filled with a stunning array of living things – animals, insects,

plants, fungi, bacteria, and more. This is apparent not just to the biologists
who study life, but also to anyone who has ever taken a walk in a park,
spent a day at a zoo, watched a nature documentary, or wondered about
the pets and pests that share a living space with us. To think about the
diversity of life in these terms – as ‘animals,’ ‘insects,’ ‘plants,’ and so on, is
to classify it. It seems to imply a division of the world into different kinds of
things – an animal kind, insect kind, and plant kind.
As anyone who has studied biology knows, modern biological classifica-
tion goes far beyond the everyday vernacular terms ‘animal,’ ‘insect,’ and
‘plant,’ employing a system based on the ideas of the Swedish botanist
Carolus Linnaeus, who developed a framework for classifying living things
(as well as minerals) that was hierarchical and comprehensive. According
to this approach, all individual organisms are grouped into species that are
then grouped together into higher level taxa – genera, orders, classes, and
kingdoms. Linnaeus also proposed a naming system based on genus and
species membership. He gave humans, for instance, the name Homo sapiens,
where the first name denotes the genus and the second name identifies the
species taxon within the genus. For Linnaeus, an individual human was
a member of the species sapiens, which was itself a member of the genus
Homo. Homo sapiens was then part of the hierarchy, by being a member of
higher level taxa – the order Anthropomorpha, class Quadrupedia, and
kingdom Animale.
The Linnaean system was adopted by Charles Darwin and given an evo-
lutionary interpretation. For Darwin, the group-in-group hierarchy of the
Linnaean system could represent the branch-on-branch structure of the

6
 Classification and the Diversity of Life 7

evolutionary tree, which in turn could represent evolutionary diversifica-

tion as new species form and diverge. Since Darwin this system has become
fully embedded in our practices and institutions – our zoos, natural history
museums, biodiversity studies, and collections. Nonetheless, the practice
of biological classification is not yet settled. There are four main ongo-
ing philosophical debates about classification. The first is theoretical: What
should a biological classification represent? Linnaeus may have thought
that classification should represent the ideas of God that governed cre-
ation, but Darwin and his followers thought biological classification should
instead have an evolutionary basis, representing genealogy and degree of
divergent change. On Darwin’s approach, which came to be known as “evo-
lutionary taxonomy” or “evolutionary systematics,” organisms should be
grouped together based on common ancestry, but the resulting taxa should
be ranked on degree of divergence. The Linnaean class Aves, for instance,
contains many species of birds, all with a common ancestor. But because
this group is so large and has undergone such great modification, it was
given an elevated taxonomic rank.
Another evolutionary approach was developed in the second half of the
twentieth century by a group of systematists known as “cladists” or “phy-
logeneticists.” They followed Darwin’s example in basing classification on
genealogy (phylogeny), but rejected the idea that ranking should be based
on degree of divergence. According to this approach, a classification should
represent only phylogeny, and more specifically, only the branching pro-
cess in evolution as new species form. This makes a difference in classifica-
tion. Cladists do not elevate Aves to a class, and instead treat birds as the
clade (branch) Avialae in Theropoda, a taxon that also includes dinosaurs
(Weishampel, Dodson, and Osmólska 2004).
But some systematists have rejected the idea that classification should
represent evolutionary history at all. In part this is motivated by the fact
that we lack precise knowledge of the phylogenetic origins of all species.
We may know, for instance, that all birds share a common ancestry, but
we don’t know the precise branching order throughout Aves and so can
reconstruct only the outline of the evolutionary tree here with any confi-
dence. Consequently we cannot classify on evolutionary grounds with cer-
tainty. Moreover, if classification represents evolutionary history and our
reconstruction of that history changes, then the classification must change
as well. What we can have though, according to these systematists, is a
8 Why Classify?

system based on the detailed analysis of similarities and differences among

taxa – a “phenetic” system. Organisms that are most similar overall get
grouped together at all levels.
The second philosophical debate is about operational procedures: How
should a classification be generated? The answer to this question is obvi-
ously dependent on the answer to the theoretical question about what a
classification should represent. Linnaeus primarily used “fructification”
traits – traits related to reproduction, thinking that was the best way to
uncover the secrets of God’s design. The Darwinians who followed, and
believed that classification should represent evolutionary history, argued
that classifications should be based only on “homologies” – shared traits
due to common ancestry, and not on “analogies,” similarities based on con-
vergent adaptive change. But how shared traits can be established as homol-
ogies has generated some controversy. Evolutionary taxonomists have typi-
cally used assumptions about evolutionary processes in general, and the
operation of natural selection in particular, to determine which shared
traits are homologies and indicate a common ancestry. Cladists (phyloge-
neticists) have disagreed, arguing that this method is circular. Assumptions
about evolutionary processes cannot be used to reconstruct the evolution-
ary past, because those process hypotheses can be confirmed only from the
reconstruction of evolutionary history. Cladists have advocated an alter-
native method based on a parsimony principle they allege to be theory
independent. According to this principle, the best hypothesis about the
evolutionary past is the one that requires the fewest assumptions of evolu-
tionary change. And in contrast to both the evolutionary taxonomists and
the phylogeneticists, those who advocate a phenetic system, based only on
similarities and differences, have typically endorsed the use of all similari-
ties and differences, rather than just those traits deemed homologies. This
makes sense because the phenetic classification was never intended to rep-
resent evolutionary history, only overall similarity.
The third philosophical debate is about the role of tree thinking in bio-
logical classification. The only diagram in the first edition of Darwin’s On
the Origin of Species was of a branching tree that represented the divergent
speciation in evolution. As Darwin used the group-in-group structure of
classification to represent the branch-on-branch structure of this tree, the
tree metaphor has permeated thinking about evolution and biological clas-
sification. Recently there have been attempts to reconstruct the one grand
 Classification and the Diversity of Life 9

tree of life. But there have also been recent challenges to tree thinking.
First, we can construct different and conflicting trees, depending on what
interests us: species taxa, organisms, character traits, or genes. The evo-
lutionary history of genes, for instance, has tree-like structures, but gene
trees often conflict with species trees. Second, while evolutionary trees typ-
ically represent only branching and diverging, there is now believed to be
substantial reticulation – the rejoining of branches, through introgression,
hybridization, or horizontal gene transfer, especially among plants, bac-
teria, and viruses. The strictly branching structure of the tree of life does
not seem to accurately represent the complicated and messy evolutionary
history. Some think we should therefore abandon this tree metaphor.
The fourth philosophical debate, about ranking, is a consequence of the
idea that classification should be based on the evolutionary tree. If biological
classification represents a branching evolutionary tree, then the Linnaean
hierarchy and naming system appear to be radically inadequate. The cur-
rent twenty levels or so of the hierarchy cannot possibly represent all the
branches of the multibillion-year-old evolutionary tree. How then can we
name and organize all the taxa? Indentation and numerical methods have
been proposed, but the Linnaean system has become so entrenched in how
we think about and represent biodiversity that it is hard to see how it could
be abandoned. Is it possible to modify the Linnaean system to better repre-
sent evolutionary history and the full diversity of life?
These philosophical debates cannot just be brushed aside. Anyone who
is engaged in the classification of living things relies, implicitly at least, on
assumptions about what should be represented and how a classification
should be constructed. This book aims to look at these issues, not from
a partisan perspective (although I have also been a participant in these
debates) but from that of a mostly impartial observer. This does not imply
that we must avoid any conclusions at all about the various claims, but it
does require that we look at them carefully and objectively. But before we
look at these issues that have engaged professional systematists about bio-
logical classification, we need to understand classification in general. After
all, modern biological classification is just one species of classification.
What is notable about biological classification is that it need not begin
with or depend on the scientific approach based on the Linnaean system.
Without consulting biologists we easily distinguish cats from dogs, bees
from spiders, birds from fish, and plants from animals. And we seemingly
10 Why Classify?

do all this naturally and in the absence of any explicit theory of classifica-
tion. But why do we classify? One initial answer is that we classify because
we must. Classification is an unavoidable natural human tendency. And
there is a tendency to classify many kinds of things, not just the living
things of our biological classifications. To understand classification in gen-
eral, we can approach it naturalistically, treating it as natural phenomena
to be studied scientifically.
A naturalistic approach reveals that classification is universal. People
in all known cultures classify living things, and in roughly similar ways.
This is hardly surprising, given what we now know about the psychology
of classification. Through observation and experiment, linguists and cog-
nitive psychologists have come to understand what seems to be an innate
and universal human tendency to classify in particular ways. In this chap-
ter, we will look first at how people in different cultures think about and
classify living things, and then at what developmental linguistics can tell
us about the psychological basis for classification. But not all classifica-
tions are equal. Some seem to reflect real divisions in the world, while
some seem arbitrary or merely pragmatic. On one standard philosophi-
cal way of thinking, we can understand this distinction in terms of the
differences between natural kinds and the merely conventional or artificial
kinds. We will briefly look at this natural kinds framework at the end of
the chapter.
A comprehensive understanding of biological classification also requires
that we know something about its history. Just as we understand human
nature partly though what we know about the evolution of Homo sapiens –
its origins in a primate lineage and its modification by natural selection and
other processes, we can understand biological classification partly through
knowledge of its origins and development. In Chapter 2 we look at what
many see as the beginning of biological classification in Aristotle’s use of
the classificatory terms ‘eidos’ (translated into Latin as ‘species’) and 'genos'
(translated as ‘genus’). We will also look at how the Aristotelian framework
was adopted and transformed in the 1,500 years after his death. In Chapter
3 we first look at the beginnings of the modern empirical approach to clas-
sification in the work of the medical herbalists and early naturalists. Then
we delve into how that approach was developed by Linnaeus and given an
evolutionary gloss by Darwin. Chapter 4 shows how Darwin’s interpreta-
tion of the Linnaean framework was further developed in the twentieth
 The Anthropology of Classification 11

century by the evolutionary taxonomists, and then how it was challenged

by pheneticists and cladists.
Tree thinking is implicit in the evolutionary approaches that take classi-
fication to represent the structure of the evolutionary tree. In Chapter 5 we
look at the various ways trees have been used, and the potential problems
with trees posed by ranking, hybridization, and horizontal gene transfer.
Chapter 6 is on what seems to be the most theoretically significant level
in classification – the species level, and the many ways of thinking about
species. Chapter 7 is on the metaphysical foundation of classification. How
should we think about the basic, fundamental nature of biological taxa?
Chapter 8 looks at the relation between evolutionary theory and classifica-
tion, and contrasts empiricist and theoretical approaches. In Chapter 9 we
conclude with what seems to be a fundamental and deep-seated tension
between the psychology of classification and the modern scientific and the-
oretical foundations of biological classification. Our psychology leads us to
think about biological classification in one way and our theories about the
world lead us in conflicting ways. As this tension lies behind many of the
philosophical debates in biological classification, we can perhaps better
understand these debates by understanding this tension.

The Anthropology of Classification

One reason to think that classification is natural is that all people seem
to do it, not just professional biologists. This is apparent in the studies
of folkbiology – how the “folk” or nonscientists think about living things,
ethnobiology and ethnotaxonomy – how the members of different cultures
think about life and its classification. What these studies seem to reveal
are broad cross-cultural similarities in the classification of life. To avoid
the bias of modern theoretical biology, ethnobiological studies have typ-
ically focused on those cultures least influenced by modern scientific
ways of thinking, from the Native American cultures of the Americas,
to a variety of relatively isolated cultures of Southeast Asia and Africa.
Typically an ethnobiologist will question an educated local “informant”
about the names and features of the living things in the local environ-
ment, hoping to discover the vernacular terms the informant applies
to these, and the implicit classificatory structure. Jared Diamond and
K. David Bishop used this method with the Ketengban people of New
12 Why Classify?

Guinea. Over a period of three weeks in 1993, they spent eight to eleven
hours a day walking through the forest with their informants, mostly
observing birds.

Our principal method for eliciting bird names consisted in asking

Ketengban guides for the name of a bird that we and they both saw, or else
heard, while walking together. In order to distinguish which individual
bird we meant if there were several in sight or calling, we either pointed
to the bird or imitated the call that we were hearing. In order to check
that the Ketengban name given in reply actually was meant to refer to the
bird about which we were inquiring, we asked our Ketengban guides to
describe the bird to us in detail – in particular, its bill, tail, size, color diet,
and forest stratum in which it is normally foraged. In that way we could
ascertain whether they and we were really talking about the same bird,
and whether they were really familiar with the species. (Diamond and
Bishop 1999, 23)

Their results:

We recorded 169 Ketengban bird names, identified most of them definitely,

and identified most others tentatively. We also recorded 127 Ketengban
names for trees, 51 names for mammals, 34 names for frogs, 16 names for
lizards, 9 names for snakes, 6 names for spiders, 4 names for butterflies,
and a few names for other insects and fungi, but we will not discuss these
other Ketengban names because we do not know the scientific identities of
most of them. (Diamond and Bishop 1999, 23)

The last sentence of this quote hints at an obvious complication.

Ethnobiologists typically approach and understand the thinking of the
local informants within the framework of their own scientifically informed
views. If they know much about Linnaean classification, as Diamond and
Bishop do, then this likely becomes the basis of the comparison. But
Diamond and Bishop also formulated their questioning to uncover ecologi-
cal or behavioral classifications that would not obviously fit into the mod-
ern, evolutionary Linnaean framework:

Gradually, as we became familiar with many Ketengban names, we

structured the questioning by asking our informants to name and describe
to us all night birds, or all grassland birds, or all ground-dwelling birds,
or all birds similar to some species (e.g., a parrot or pigeon species) whose
vernacular name we had already identified. (Diamond and Bishop 1999, 31)
 The Anthropology of Classification 13

Notice that Grassland birds is a hybrid category, based partly on ecology, and
groups together birds that are not necessarily closely related in evolution-
ary terms. Similarly, night birds and ground-dwelling birds are partly behav-
ioral and ecological groupings. So none of these taxa would fit into the
modern evolutionary Linnaean framework based on common ancestry.
The most obvious thing to notice about the findings of Diamond and
Bishop is that the Ketengban informants have specific names for, and know
a great deal more about the living things in their environment than do
the average members of modern, scientific societies. This is unsurprising
since their daily life depends much more on knowledge of the animals
and plants in their environment than does life in urban and suburban cul-
tures. But another notable conclusion, according to Diamond and Bishop,
is that the vernacular Ketengban bird names seem to refer to the bird spe-
cies recognized by scientists, with just a few exceptions where a group
of related species might be given a single name, or a sexually dimorphic
species might have different names for the female and male (Diamond and
Bishop 1999, 35–38).
Diamond and Bishop also looked at the hierarchical structure of the
Ketengban classification, and compared it to the Linnaean system:

Scientific nomenclature for a local biota is hierarchical, with four major

levels below the class level (birds being the class Aves). Those four levels
are the order, family, genus, and species. In contrast, Ketengban names
belong to only two levels: a low-level terminal category corresponding
closely to species, and a high-level collective category corresponding
approximately to classes or orders. The six collective Ketengban names
that we obtained correspond respectively to birds, bats, mammals
other than bats, snakes, lizards, and frogs. We found no evidence that
Ketengbans name any category intermediate between the low-level
terminal category and their high-level collective category. Even though
Ketengbans readily understood our questions about naming all species
in distinctive bird families, such as naming all parrot species or all
hawk species, they offered no name for those intermediate categories
(which scientists recognize as families or orders), despite their ability
to grasp the bounds of the intermediate category. (Diamond and
Bishop 1999, 32)

With the Ketengban there is a group-in-group structure, as in the Linnaean

system. The lower level categories are grouped within increasingly higher
14 Why Classify?

level categories. And while the Ketengbans have names for only two levels
of classification, they seem to recognize intermediate levels between the
higher collective level and the lower terminal level.
Studies of the folk classifications of other cultures have arrived at
roughly similar conclusions. First, these studies tend to find a relatively
large number of names and categories of plants and animals. Brent
Berlin’s study of the Tzeltal Mayan’s classification, for instance, found
hundreds of names and categories of plants (Berlin 1999). And Scott Atran
found hundreds of recognized categories of snakes, birds, and palms
among the Itzaj Mayans (Atran 1999). Second, these studies tend to find
a hierarchical structure, usually more complex than what Diamond and
Bishop found in the Ketengbans. According to Berlin, there are up to six
levels or ranks in the hierarchies he has studied. The highest is kingdom,
followed by life form, intermediate, generic, specific, and varietal (Berlin 1992,
22). Not all of these hierarchical levels would necessarily be found in all
cultures though. Nor will all the levels necessarily have a name (Berlin
1992, 31–33). Ralph Bulmer, in his studies of Kalam ethnotaxonomy, iden-
tified five levels of hierarchy, from the highest, primary taxa, which are
not subsumable into any larger taxon, to the terminal taxa, which have no
named subdivisions (Berlin 1992, 65). Third, there also seems to be a basic
privileged level in the hierarchy. According to Berlin this is the generic
level. Most of the taxa are found here. These taxa are easily recognizable
and have simple names (such as ‘dog’ and ‘cat’ in the vernacular English;
Berlin 1992, 64). Atran dubbed the groupings at this level ‘generic species’
(Atran 1999, 124).
What generalizations can we draw from these studies in the folkbiology
of various cultures? First, all folk classifications seem to be hierarchical,
with a group-in-group structure. Although Diamond and Bishop found only
two explicitly named levels in their investigation of Ketengban folk classi-
fication, more levels were implicitly recognized. Berlin, Bulmer, and Atran
have all found more, albeit not always named, levels. Atran argued that
there is a default hierarchy comprising five levels. According to Atran, the
highest level is kingdom (Atran 1999, 122). Folk kingdoms, if not explicitly
named, may be indicated by the use of a particular suffix or term. The Itzaj
used a particular term only for plants, for instance. They also used a term
translated as ‘forest-thing’ at the kingdom level that includes many verte-
brates, invertebrates, birds, and fish (Atran 1999, 123). The next highest
 The Anthropology of Classification 15

default level is life form, a level that is often based on functional, develop-
mental, or ecological factors:

The majority of taxa of lesser rank fall under one or another life form.
Most life-form taxa are named by lexically unanalyzable names (primary
lexemes), and they have further subdivisions such as tree or bird.
Biologically, members of a single life form are diverse . . . Life-form taxa
may represent general adaptations to broad sets of ecological conditions,
such as the competition of single-stem plants for sunlight and tetrapod
adaptation to life in the air. (Atran 1999, 122)

Among the Itzaj Maya, life forms in the plant kingdom include trees, shrubs,
vines, and grasses. Animal life forms include categories roughly correspond-
ing to mammals (excluding bats), birds (with bats), and herpetofauna
(amphibians and reptiles) (Atran 1999, 123, fn. 5).
The third highest level is the generic species, usually identified as species-
like groupings. This level seems to have a special status, and for a variety of
reasons, linguistic, inferential, psychological, and developmental:

The rank of generic species is the level at which morphological, behavioral

and ecological relationships between organisms maximally covary. The
majority of Itzaj folkbiological taxa belong to this level. It is this level
that Itzaj privilege when they see and talk about biological continuities.
Generic species represent cuts in nature that Itzaj children first name and
form an image of . . . and that Itzaj adults most frequently use in speech,
most easily recall in memory, and most readily communicate to others . . .
It is the rank at which Itzaj, like other folk around the world, are most
likely to attribute biological properties, including characteristic patterns
of inheritance, growth, physiological function, as well as more “hidden”
properties such as hitherto unknown organic processes, organs and
diseases. (Atran 1999, 127)

The level of generic species seems to be the primary focus of how the Itzaj
and other folk talk and think about living things. It is generic species that
are most easy to identify, name, remember, and think about. And it is at
this level that there is the strongest tendency to generalize: from the fact
that some individuals of a generic species have some trait, other individu-
als must also have that trait. Generic species also typically have simple
names, such as we see with the English vernacular terms ‘dog,’ ‘cat,’ ‘oak’
and ‘robin.’
16 Why Classify?

The lowest levels are the folkspecific and folkvarietal. Some of the most cul-
turally significant generic species get subdivided into folkspecifics that are
distinguished in part by their names. Generic species tend to have simple
names, whereas folkspecifics tend to have binomials constructed out of
the generic species name, such as we see with ‘white oak’ and ‘mountain
robin.’ At the lowest level is the folkvarietal. This level is often indicated by a
trinomial, as in ‘spotted white oak’ versus the ‘swamp white oak,’ and ‘yel-
low village papaya’ as distinguished from the ‘white village papaya.’ The
Itzaj typically distinguished folkvarietals among those living things that
have significant perceptual differences (color perhaps) and specific practi-
cal uses or cultural significance (Atran 1999, 127–128).
Many who study these folk classifications claim there is a standard
folk level of classification corresponding to the modern scientific species
level. Atran claims, for instance: “Humans everywhere classify animals and
plants into species like groupings that are as obvious to a modern scientist
as to Maya Indian” (Atran 1999, 120). And Diamond and Bishop, as quoted
earlier, claim that for the Ketengban there is a “low-level terminal category
corresponding closely to species” (Diamond and Bishop 1999, 32). Diamond
and Bishop argue further that this correspondence is evidence for realism
about species taxa.

If peoples with very different upbringings and motivations for naming

nevertheless tended to recognize the same units of Nature, that would lend
support to the view that those units correspond to an objective reality that
is not a mere invention of scientists. (Diamond and Bishop 1999, 17)

If people all over the world and in different cultures divide up biodiver-
sity into species-like groupings in the same way, then there must be some
objective, human independent significance to the units of grouping and
division. There really must be these species-like groupings in nature!
But there are problems here. First, as we shall see in Chapter 6, there is
substantial disagreement among modern systematists about the nature of
species, and consequently how species taxa are identified and individuated.
Some systematists appeal to reproductive cohesion and isolation, others to
morphological, behavior or genetic similarity, and yet others to the fact
that species are lineages. This leads to disagreement about species counts –
the number of species taxa represented by a group of organisms. If so,
there is no single set of species taxa recognized by all scientifically trained
 The Anthropology of Classification 17

systematists, and we should be skeptical of the claim that folk classifica-

tions recognize precisely the same species taxa every modern systematist
recognizes. The claim that the folk recognize the same species as modern
scientists wrongly assumes agreement among scientists.
Second, many of the scientific species concepts are “dimensional” in
the sense that they conceive of species taxa extending in space as popu-
lations, and over time as lineages. And within the populations and lin-
eages, scientific thinking about species allows for substantial variation.
This is a consequence of the fact that scientific thinking about species
taxa is informed by evolutionary theory. But just because a culture names
a kind of living thing that is part of a population and lineage does not
imply that the members of the culture think of that kind of living thing
in these ways – dimensionally as populations and lineages. The naming
of a living organism as a kind of thing is compatible with many different
conceptions of the kind itself. The folk might be thinking of the species-
like category in a variety of ways, from various kinds of similarity, but
also perhaps on religious or mystical grounds, in ways not shared with
modern systematists. The folk informant and the ethnobiologist may
therefore agree that a small set of organisms may represent the same
number of groupings, but have different ideas about what that means in
terms of the theoretical basis of the category or overall hierarchy. These
differences may affect how new specimens might get classified. One can-
not therefore conclude from the same limited number of groupings that
the classifications are the same.
Third, these species-like groupings are not necessarily at the species
level of the Linnaean classification, as Atran acknowledges:

Generic species often correspond to scientific genera or species, at least

for those organisms that humans mostly readily perceive, such as large
vertebrates and flowering plants. On occasion, generic species correspond
to local fragments of biological families (e.g., vulture), orders (e.g., bat), and
especially with invertebrates, higher-order taxa. (Atran 1999, 125)

The Linnaean classificatory levels don’t get distinguished in folk biology

because, as Atran points out, many of the genera in any particular place
are monospecific – represented by only a single species. If so, there would
be no reason to distinguish the species level from a higher level in the clas-
sification. A single folk grouping may then be ambiguous between species
18 Why Classify?

and genus levels. Moreover, when there are multiple species of a single
genus, these species may be difficult to distinguish perceptually. There may
be too little of a morphological gap for the folk taxonomists to distinguish
the two species (Atran 1999, 125–126).
Lurking behind these three problems is a more general philosophical
problem – how to determine the mapping of a term in one classificatory sys-
tem onto another system. When a proper name is applied to an individual
object, the mapping of the name is relatively unproblematic. A particular
horse can be dubbed ‘Secretariat’ and the reference of the name is clear –
the name refers to a particular horse, but not to any other horse, not to
any other creature, nor to any other object. But the mapping of a general
term such as ‘horse’ to things in the world is more problematic. We can
apply the term to a particular individual horse, perhaps the one named
‘Secretariat,’ but it isn’t necessarily clear to what other objects the term
‘horse’ would apply.
This problem can perhaps be best understood through an example.
Suppose a culture has experience and knowledge of only a single species
(in the modern scientific sense) of horse, Equus caballus, and the members
of the culture use the term ‘horse’ on the basis of this experience to refer
to particular horses they experience and that are of this species. But why
should we interpret the use of the term ‘horse’ to refer to just living things
in the scientific species taxon Equus caballus rather than living things in
the genus Equus or family Equidae? The individual horses are just as much
members of the genus and family as they are the species. In other words,
why should we take the vernacular term ‘horse’ to refer just to things of
a particular species, and therefore at the species level? The people who
use the vernacular term do not make any of the distinctions between the
various species of the genus Equus or the genera of the family Equidae.
Whether they could and would make the distinction between things of dif-
ferent species, and what terms they would use is an open question.
There are at least two options here. First, we could claim that ‘horse’
maps onto one of the three taxanomic levels – Equus caballus, Equus, or
Equidae. If so, then we would need to give reasons why one mapping is bet-
ter than other. Moreover, we would need to give a reason why it shouldn’t
map at an even higher level. Second, we could just conclude that the local
vernacular term ‘horse’ is indeterminate relative to the Linnaean system,
and cannot be identified with a particular taxon or classificatory level.
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of The stereoscope
This ebook is for the use of anyone anywhere in the United
States and most other parts of the world at no cost and with
almost no restrictions whatsoever. You may copy it, give it away
or re-use it under the terms of the Project Gutenberg License
included with this ebook or online at www.gutenberg.org. If you
are not located in the United States, you will have to check the
laws of the country where you are located before using this
eBook.

Title: The stereoscope

its history, theory, and construction, with its application
to the fine and useful arts and to education

Author: David Brewster

Release date: August 24, 2023 [eBook #71483]

Language: English

Original publication: London: John Murray, 1856

Credits: deaurider and the Online Distributed Proofreading Team

at https://www.pgdp.net (This file was produced from
images generously made available by The Internet
Archive)

*** START OF THE PROJECT GUTENBERG EBOOK THE

STEREOSCOPE ***
 THE STEREOSCOPE
ITS HISTORY, THEORY, AND
CONSTRUCTION
WITH ITS APPLICATION TO THE FINE AND
USEFUL ARTS
AND TO EDUCATION.

BY
SIR DAVID BREWSTER,
K.H., D.C.L., F.R.S., M.R.I.A.,
VICE-PRESIDENT OF THE ROYAL SOCIETY OF EDINBURGH,
ONE OF THE EIGHT
ASSOCIATES OF THE IMPERIAL INSTITUTE OF FRANCE,
OFFICER OF THE
LEGION OF HONOUR, CHEVALIER OF THE PRUSSIAN ORDER
OF MERIT,
HONORARY OR CORRESPONDING MEMBER OF THE
ACADEMIES
OF PETERSBURGH, VIENNA, BERLIN, COPENHAGEN,
STOCKHOLM, BRUSSELS, GÖTTINGEN, MODENA,
AND OF THE NATIONAL INSTITUTE OF
WASHINGTON, ETC.

WITH FIFTY WOOD ENGRAVINGS.

 LONDON:
JOHN MURRAY, ALBEMARLE STREET.
1856.
[The Right of Translation is reserved.]
EDINBURGH:
T. CONSTABLE, PRINTER TO HER MAJESTY.
 CONTENTS
Page
Introduction, 1
Chap. I.—History of the Stereoscope, 5
II.—On Monocular Vision, or Vision with One Eye, 38
III.—On Binocular Vision, or Vision with Two Eyes, 47
IV.—Description of the Ocular, Reflecting,
and Lenticular Stereoscopes, 53
V.—On the Theory of the Stereoscopic Vision, 76
VI.—On the Union of Similar Pictures in Binocular Vision 90
VII.—Description of different Stereoscopes, 107
VIII.—Method of taking Pictures for the Stereoscope, 131
IX.—On the Adaptation of the Pictures to the Stereoscope.
—Their Size, Position, and Illumination 159
X.—Application of the Stereoscope to Painting, 166
XI.—Application of the Stereoscope to Sculpture,
Architecture, and Engineering, 183
XII.—Application of the Stereoscope to Natural History, 189
XIII.—Application of the Stereoscope to Educational Purposes, 193
XIV.—Application of the Stereoscope to Purposes of Amusement, 204
XV.—On the Production of Stereoscopic Pictures
from a Single Picture, 211
XVI.—On certain Fallacies of Sight in theVision of Solid Bodies, 216
XVII.—On certain Difficulties experienced in the Use of
the Stereoscope, 231
ON THE STEREOSCOPE.
 INTRODUCTION.
The Stereoscope, a word derived from στέρεος, solid, and
σκόπειν, to see, is an optical instrument, of modern invention, for
representing, in apparent relief and solidity, all natural objects and all
groups or combinations of objects, by uniting into one image two
plane representations of these objects or groups as seen by each
eye separately. In its most general form the Stereoscope is a
binocular instrument, that is, is applied to both eyes; but in two of its
forms it is monocular, or applied only to one eye, though the use of
the other eye, without any instrumental aid, is necessary in the
combination of the two plane pictures, or of one plane picture and its
reflected image. The Stereoscope, therefore, cannot, like the
telescope and microscope, be used by persons who have lost the
use of one eye, and its remarkable effects cannot be properly
appreciated by those whose eyes are not equally good.
When the artist represents living objects, or groups of them, and
delineates buildings or landscapes, or when he copies from statues
or models, he produces apparent solidity, and difference of distance
from the eye, by light and shade, by the diminished size of known
objects as regulated by the principles of geometrical perspective,
and by those variations in distinctness and colour which constitute
what has been called aerial perspective. But when all these
appliances have been used in the most skilful manner, and art has
exhausted its powers, we seldom, if ever, mistake the plane picture
for the solid which it represents. The two eyes scan its surface, and
by their distance-giving power indicate to the observer that every
point of the picture is nearly at the same distance from his eye. But if
the observer closes one eye, and thus deprives himself of the power
of determining differences of distance by the convergency of the
optical axes, the relief of the picture is increased. When the pictures
are truthful photographs, in which the variations of light and shade
are perfectly represented, a very considerable degree of relief and
solidity is thus obtained; and when we have practised for a while this
species of monocular vision, the drawing, whether it be of a statue, a
living figure, or a building, will appear to rise in its different parts from
the canvas, though only to a limited extent.
In these observations we refer chiefly to ordinary drawings held in
the hand, or to portraits and landscapes hung in rooms and galleries,
where the proximity of the observer, and lights from various
directions, reveal the surface of the paper or the canvas; for in
panoramic and dioramic representations, where the light, concealed
from the observer, is introduced in an oblique direction, and where
the distance of the picture is such that the convergency of the optic
axes loses much of its distance-giving power, the illusion is very
perfect, especially when aided by correct geometrical and aerial
perspective. But when the panorama is illuminated by light from
various directions, and the slightest motion imparted to the canvas,
its surface becomes distinctly visible, and the illusion instantly
disappears.
The effects of stereoscopic representation are of a very different
kind, and are produced by a very different cause. The singular relief
which it imparts is independent of light and shade, and of
geometrical as well as of aerial perspective. These important
accessories, so necessary in the visual perception of the drawings in
plano, avail nothing in the evolution of their relievo, or third
dimension. They add, doubtless, to the beauty of the binocular
pictures; but the stereoscopic creation is due solely to the
superposition of the two plane pictures by the optical apparatus
employed, and to the distinct and instantaneous perception of
distance by the convergency of the optic axes upon the similar points
of the two pictures which the stereoscope has united.
If we close one eye while looking at photographic pictures in the
stereoscope, the perception of relief is still considerable, and
approximates to the binocular representation; but when the pictures
are mere diagrams consisting of white lines upon a black ground, or
black lines upon a white ground, the relief is instantly lost by the
shutting of the eye, and it is only with such binocular pictures that we
see the true power of the stereoscope.
 As an amusing and useful instrument the stereoscope derives
much of its value from photography. The most skilful artist would
have been incapable of delineating two equal representations of a
figure or a landscape as seen by two eyes, or as viewed from two
different points of sight; but the binocular camera, when rightly
constructed, enables us to produce and to multiply photographically
the pictures which we require, with all the perfection of that
interesting art. With this instrument, indeed, even before the
invention of the Daguerreotype and the Talbotype, we might have
exhibited temporarily upon ground-glass, or suspended in the air, the
most perfect stereoscopic creations, by placing a Stereoscope
behind the two dissimilar pictures formed by the camera.
 CHAPTER I.
HISTORY OF THE STEREOSCOPE.

When we look with both eyes open at a sphere, or any other solid
object, we see it by uniting into one two pictures, one as seen by the
right, and the other as seen by the left eye. If we hold up a thin book
perpendicularly, and midway between both eyes, we see distinctly
the back of it and both sides with the eyes open. When we shut the
right eye we see with the left eye the back of the book and the left
side of it, and when we shut the left eye we see with the right eye the
back of it and the right side. The picture of the book, therefore, which
we see with both eyes, consists of two dissimilar pictures united,
namely, a picture of the back and the left side of the book as seen by
the left eye, and a picture of the back and right side of the book as
seen by the right eye.
In this experiment with the book, and in all cases where the
object is near the eye, we not only see different pictures of the same
object, but we see different things with each eye. Those who wear
spectacles see only the left-hand spectacle-glass with the left eye,
on the left side of the face, while with the right eye they see only the
right-hand spectacle-glass on the right side of the face, both glasses
of the spectacles being seen united midway between the eyes, or
above the nose, when both eyes are open. It is, therefore, a fact well
known to every person of common sagacity that the pictures of
bodies seen by both eyes are formed by the union of two dissimilar
pictures formed by each.
This palpable truth was known and published by ancient
mathematicians. Euclid knew it more than two thousand years ago,
as may be seen in the 26th, 27th, and 28th theorems of his Treatise
on Optics.[1] In these theorems he shews that the part of a sphere
seen by both eyes, and having its diameter equal to, or greater or
less than the distance between the eyes, is equal to, and greater or
less than a hemisphere; and having previously shewn in the 23d and
24th theorems how to find the part of any sphere that is seen by one
eye at different distances, it follows, from constructing his figure, that
each eye sees different portions of the sphere, and that it is seen by
both eyes by the union of these two dissimilar pictures.
More than fifteen hundred years ago, the celebrated physician
Galen treated the subject of binocular vision more fully than Euclid.
In the twelfth chapter of the tenth book of his work, On the use of the
different parts of the Human Body, he has described with great
minuteness the various phenomena which are seen when we look at
bodies with both eyes, and alternately with the right and the left. He
shews, by diagrams, that dissimilar pictures of a body are seen in
each of these three modes of viewing it; and, after finishing his
demonstration, he adds,—
“But if any person does not understand these
demonstrations by means of lines, he will finally give
his assent to them when he has made the following
experiment:—Standing near a column, and shutting
each of the eyes in succession;—when the right eye is
shut, some of those parts of the column which were
previously seen by the right eye on the right side of the
column, will not now be seen by the left eye; and when
the left eye is shut, some of those parts which were
formerly seen by the left eye on the left side of the
column, will not now be seen by the right eye. But
when we, at the same time, open both eyes, both
these will be seen, for a greater part is concealed
when we look with either of the two eyes, than when
we look with both at the same time.”[2]
In such distinct and unambiguous terms, intelligible to the
meanest capacity, does this illustrious writer announce the
fundamental law of binocular vision—the grand principle of the
Stereoscope, namely, that the picture of the solid column which we
see with both eyes is composed of two dissimilar pictures, as seen
by each eye separately. As the vision of the solid column, therefore,
was obtained by the union of these dissimilar pictures, an instrument
only was wanted to take such pictures, and another to combine
them. The Binocular Photographic Camera was the one instrument,
and the Stereoscope the other.
The subject of binocular vision was studied by various optical
writers who have flourished since the time of Galen. Baptista Porta,
one of the most eminent of them, repeats, in his work On Refraction,
the propositions of Euclid on the vision of a sphere with one and both
eyes, and he cites from Galen the very passage which we have
given above on the dissimilarity of the three pictures seen by each
eye and by both. Believing that we see only with one eye at a time,
he denies the accuracy of Euclid’s theorems, and while he admits
the correctness of the observations of Galen, he endeavours to
explain them upon other principles.

Fig. 1.
In illustrating the views of Galen on the dissimilarity of the three
pictures which are requisite in binocular vision, he employs a much
more distinct diagram than that which is given by the Greek
physician. “Let a,” he says, “be the pupil of the right eye, b that of the
left, and dc the body to be seen. When we look at the object with
both eyes we see dc, while with the left eye we see ef, and with the
right eye gh. But if it is seen with one eye, it will be seen otherwise,
for when the left eye b is shut, the body cd, on the left side, will be
seen in hg; but when the right eye is shut, the body cd will be seen
in fe, whereas, when both eyes are opened at the same time, it will
be seen in cd.” These results are then explained by copying the
passage from Galen, in which he supposes the observer to repeat
these experiments when he is looking at a solid column.
In looking at this diagram, we recognise at once not only the
principle, but the construction of the stereoscope. The double
stereoscopic picture or slide is represented by he; the right-hand
picture, or the one seen by the right eye, by hf; the left-hand picture,
or the one seen by the left eye, by ge; and the picture of the solid
column in full relief by dc, as produced midway between the other
two dissimilar pictures, hf and ge, by their union, precisely as in the
stereoscope.[3]
Galen, therefore, and the Neapolitan philosopher, who has
employed a more distinct diagram, certainly knew and adopted the
fundamental principle of the stereoscope; and nothing more was
required, for producing pictures in full relief, than a simple instrument
for uniting hf and ge, the right and left hand dissimilar pictures of the
column.
 Fig. 2.
In the treatise on painting which he left behind him in MS.,[4]
Leonardo da Vinci has made a distinct reference to the dissimilarity
of the pictures seen by each eye as the reason why “a painting,
though conducted with the greatest art, and finished to the last
perfection, both with regard to its contours, its lights, its shadows,
and its colours, can never shew a relievo equal to that of the natural
objects, unless these be viewed at a distance and with a single
eye,”[5] which he thus demonstrates. “If an object c be viewed by a
single eye at a, all objects in the space behind it—included, as it
were, in a shadow ecf, cast by a candle at a—are invisible to an eye
at a; but when the other eye at b is opened, part of these objects
become visible to it; those only being hid from both eyes that are
included, as it were, in the double shadow cd, cast by two lights at a
and b and terminated in d; the angular space edg, beyond d, being
always visible to both eyes. And the hidden space cd is so much the
shorter as the object c is smaller and nearer to the eyes. Thus he
observes that the object c, seen with both eyes, becomes, as it
were, transparent, according to the usual definition of a transparent
thing, namely, that which hides nothing beyond it. But this cannot
happen when an object, whose breadth is bigger than that of the
pupil, is viewed by a single eye. The truth of this observation is,
therefore, evident, because a painted figure intercepts all the space
behind its apparent place, so as to preclude the eyes from the sight
of every part of the imaginary ground behind it. Hence,” continues
Dr. Smith, “we have one help to distinguish the place of a near object
more accurately with both eyes than with one, inasmuch as we see it
more detached from other objects beyond it, and more of its own
surface, especially if it be roundish.”
We have quoted this passage, not from its proving that Leonardo
da Vinci was acquainted with the fact that each eye, a, b, sees
dissimilar pictures of the sphere c, but because it has been referred
to by Mr. Wheatstone as the only remark on the subject of binocular
vision which he could find “after looking over the works of many
authors who might be expected to have made them.” We think it
quite clear, however, that the Italian artist knew as well as his
commentator Dr. Smith, that each eye, a and b, sees dissimilar parts
of the sphere c. It was not his purpose to treat of the binocular
pictures of c, but his figure proves their dissimilarity.
The subject of binocular vision was successfully studied by
Francis Aguillon or Aguilonius,[6] a learned Jesuit, who published his
Optics in 1613. In the first book of his work, where he is treating of
the vision of solids of all forms, (de genere illorum quæ τὰ στέρεα [ta
sterea] nuncupantur,) he has some difficulty in explaining, and fails
to do it, why the two dissimilar pictures of a solid, seen by each eye,
do not, when united, give a confused and imperfect view of it. This
discussion is appended to the demonstration of the theorem, “that
when an object is seen with two eyes, two optical pyramids are
formed whose common base is the object itself, and whose vertices
are in the eyes,”[7] and is as follows:—
“When one object is seen with two eyes, the angles at the
vertices of the optical pyramids (namely, haf, gbe, Fig. 1) are not
always equal, for beside the direct view in which the pyramids ought
to be equal, into whatever direction both eyes are turned, they
receive pictures of the object under inequal angles, the greatest of
which is that which is terminated at the nearer eye, and the lesser
that which regards the remoter eye. This, I think, is perfectly evident;
but I consider it as worthy of admiration, how it happens that bodies
seen by both eyes are not all confused and shapeless, though we
view them by the optical axes fixed on the bodies themselves. For
greater bodies, seen under greater angles, appear lesser bodies
under lesser angles. If, therefore, one and the same body which is in
reality greater with one eye, is seen less on account of the inequality
of the angles in which the pyramids are terminated, (namely, haf,
gbe,[8]) the body itself must assuredly be seen greater or less at the
same time, and to the same person that views it; and, therefore,
since the images in each eye are dissimilar (minime sibi congruunt)
the representation of the object must appear confused and disturbed
(confusa ac perturbata) to the primary sense.”
“This view of the subject,” he continues, “is certainly consistent
with reason, but, what is truly wonderful is, that it is not correct, for
bodies are seen clearly and distinctly with both eyes when the optic
axes are converged upon them. The reason of this, I think, is, that
the bodies do not appear to be single, because the apparent images,
which are formed from each of them in separate eyes, exactly
coalesce, (sibi mutuo exacte congruunt,) but because the common
sense imparts its aid equally to each eye, exerting its own power
equally in the same manner as the eyes are converged by means of
their optical axes. Whatever body, therefore, each eye sees with the
eyes conjoined, the common sense makes a single notion, not
composed of the two which belong to each eye, but belonging and
accommodated to the imaginative faculty to which it (the common
sense) assigns it. Though, therefore, the angles of the optical
pyramids which proceed from the same object to the two eyes,
viewing it obliquely, are inequal, and though the object appears
greater to one eye and less to the other, yet the same difference
does not pass into the primary sense if the vision is made only by the
axes, as we have said, but if the axes are converged on this side or
on the other side of the body, the image of the same body will be
seen double, as we shall shew in Book iv., on the fallacies of vision,
and the one image will appear greater and the other less on account
of the inequality of the angles under which they are seen.”[9]
 Such is Aguilonius’s theory of binocular vision, and of the union
of the two dissimilar pictures in each eye by which a solid body is
seen. It is obviously more correct than that of Dr. Whewell and Mr.
Wheatstone. Aguilonius affirms it to be contrary to reason that two
dissimilar pictures can be united into a clear and distinct picture, as
they are actually found to be, and he is therefore driven to call in the
aid of what does not exist, a common sense, which rectifies the
picture. Dr. Whewell and Mr. Wheatstone have cut the Gordian knot
by maintaining what is impossible, that in binocular and stereoscopic
vision a long line is made to coincide with a short one, and a large
surface with a small one; and in place of conceiving this to be done
by a common sense overruling optical laws, as Aguilonius supposes,
they give to the tender and pulpy retina, the recipient of ocular
pictures, the strange power of contracting or expanding itself in order
to equalize inequal lines and inequal surfaces!

Fig. 3.
In his fourth and very interesting book, on the fallacies of
distance, magnitude, position, and figure, Aguilonius resumes the
subject of the vision of solid bodies. He repeats the theorems of
Euclid and Gassendi on the vision of the sphere, shewing how much
of it is seen by each eye, and by both, whatever be the size of the
sphere, and the distance of the observer. At the end of the theorems,
in which he demonstrates that when the diameter of the sphere is
equal to the distance between the eyes we see exactly a
hemisphere, he gives the annexed drawing of the mode in which the
sphere is seen by each eye, and by both. In this diagram e is the
right eye and d the left, chfi the section of that part of the sphere bc
which is seen by the right eye e, bhga the section of the part which
is seen by the left eye d, and blc the half of the great circle which is
the section of the sphere as seen by both eyes.[10] These three
pictures of the solids are all dissimilar. The right eye e does not see
the part blcif of the sphere; the left eye does not see the part
blcga, while the part seen with both eyes is the hemisphere blcgf,
the dissimilar segments bfg, cgf being united in its vision.[11]
After demonstrating his theorems on the vision of spheres with
one and both eyes,[12] Aguilonius informs us, before he proceeds to
the vision of cylinders, that it is agreed upon that it is not merely true
with the sphere, but also with the cylinder, the cone, and all bodies
whatever, that the part which is seen is comprehended by tangent
rays, such as eb, ec for the right eye, in Fig. 3. “For,” says he, “since
these tangent lines are the outermost of all those which can be
drawn to the proposed body from the same point, namely, that in
which the eye is understood to be placed, it clearly follows that the
part of the body which is seen must be contained by the rays
touching it on all sides. For in this part no point can be found from
which a right line cannot be drawn to the eye, by which the correct
visible form is brought out.”[13]
Optical writers who lived after the time of Aguilonius seem to
have considered the subject of binocular vision as exhausted in his
admirable work. Gassendi,[14] though he treats the subject very
slightly, and without any figures, tells us that we see the left side of
the nose with the left eye, and the right side of it with the right eye,—
two pictures sufficiently dissimilar. Andrew Tacquet,[15] though he
quotes Aguilonius and Gassendi on the subject of seeing distances
with both eyes, says nothing on the binocular vision of solids; and
Smith, Harris, and Porterfield, only touch upon the subject
incidentally. In commenting on the passage which we have already
quoted from Leonardo da Vinci, Dr. Smith says, “Hence we have one
help to distinguish the place of a near object more accurately with
both eyes than with one, inasmuch as we see it more detached from
other objects beyond it, and more of its own surface, especially if it
be roundish.”[16] If any farther evidence were required that Dr. Smith
was acquainted with the dissimilarity of the images of a solid seen by
each eye, it will be found in his experiment with a “long ruler placed
between the eyebrows, and extended directly forward with its flat
sides, respecting the right hand and the left.” “By directing the eyes
to a remote object,” he adds, “the right side of the ruler seen by the
right eye will appear on the left hand, and the left side on the right
hand, as represented in the figure.”[17]
In his Treatise on Optics, published in 1775, Mr. Harris, when
speaking of the visible or apparent figures of objects, observes, that
“we have other helps for distinguishing prominences of small parts
besides those by which we distinguish distances in general, as their
degrees of light and shade, and the prospect we have round them.”
And by the parallax, on account of the distance betwixt our eyes, we
can distinguish besides the front part of the two sides of a near
object not thicker than the said distance, and this gives a visible
relievo to such objects, which helps greatly to raise or detach them
from the plane in which they lie. Thus the nose on a face is the more
remarkably raised by our seeing both sides of it at once.“[18] That is,
the relievo is produced by the combination of the two dissimilar
pictures given by each eye.
Without referring to a figure given by Dr. Porterfield, in which he
actually gives drawings of an object as seen by each eye in
binocular vision,[19] the one exhibiting the object as seen endwise by
the right eye, and the other the same object as seen laterally by the
left eye, we may appeal to the experience of every optical, or even of
every ordinary observer, in support of the fact, that the dissimilarity of
the pictures in each eye, by which we see solid objects, is known to
those who have never read it in Galen, Porta, or Aguilonius. Who
has not observed the fact mentioned by Gassendi and Harris, that
their left eye sees only the left side of their nose, and their right eye
the right side, two pictures sufficiently dissimilar? Who has not
noticed, as well as Dr. Smith, that when they look at any thin, flat
body, such as a thin book, they see both sides of it—the left eye only
the left side of it, and the right eye only the right side, while the back,
or the part nearest the face, is seen by each eye, and both the sides
and the back by both the eyes? What student of perspective is there
—master or pupil, male or female—who does not know, as certainly
as he knows his alphabet, that the picture of a chair or table, or
anything else, drawn from one point of sight, or as seen by one eye
placed in that point, is necessarily dissimilar to another drawing of
the same object taken from another point of sight, or as seen by the
other eye placed in a point 2½ inches distant from the first? If such a
person is to be found, we might then admit that the dissimilarity of
the pictures in each eye was not known to every student of
perspective.[20]
Such was the state of our knowledge of binocular vision when
two individuals, Mr. Wheatstone, and Mr. Elliot, now Teacher of
Mathematics in Edinburgh, were directing their attention to the
subject. Mr. Wheatstone communicated an important paper on the
Physiology of Vision to the British Association at Newcastle in
August 1838, and exhibited an instrument called a Stereoscope, by
which he united the two dissimilar pictures of solid bodies, the τὰ
στέρεα, (ta sterea of Aguilonius,) and thus reproduced, as it were,
the bodies themselves. Mr. Wheatstone’s paper on the subject,
which had been previously read at the Royal Society on the 21st of
June, was printed in their Transactions for 1838.[21]
Mr. Elliot was led to the study of binocular vision in consequence
of having written an Essay, so early as 1823, for the Class of Logic in
the University of Edinburgh, “On the means by which we obtain our
knowledge of distances by the Eye.” Ever since that date he was
familiar with the idea, that the relief of solid bodies seen by the eye
was produced by the union of the dissimilar pictures of them in each
eye, but he never imagined that this idea was his own, believing that
it was known to every student of vision. Previous to or during the
year 1834, he had resolved to construct an instrument for uniting two
dissimilar pictures, or of constructing a stereoscope; but he delayed
doing this till the year 1839, when he was requested to prepare an
original communication for the Polytechnic Society, which had been
recently established in Liverpool. He was thus induced to construct
the instrument which he had projected, and he exhibited it to his
friends, Mr. Richard Adie, optician, and Mr. George Hamilton,
lecturer on chemistry in Liverpool, who bear testimony to its
existence at that date. This simple stereoscope, without lenses or
mirrors, consisted of a wooden box 18 inches long, 7 broad, and 4½
deep, and at the bottom of it, or rather its farther end, was placed a
slide containing two dissimilar pictures of a landscape as seen by
each eye. Photography did not then exist, to enable Mr. Elliot to
procure two views of the same scene, as seen by each eye, but he
drew the transparency of a landscape with three distances. The first
and most remote was the moon and the sky, and a stream of water
from which the moon was reflected, the two moons being placed
nearly at the distance of the two eyes, or 2½ inches, and the two
reflected moons at the same distance. The second distance was
marked by an old cross about a hundred feet off; and the third
distance by the withered branch of a tree, thirty feet from the
observer. In the right-hand picture, one arm of the cross just touched
the disc of the moon, while, in the left-hand one, it projected over
one-third of the disc. The branch of the tree touched the outline of a
distant hill in the one picture, but was “a full moon’s-breadth” from it
on the other. When these dissimilar pictures were united by the eyes,
a landscape, certainly a very imperfect one, was seen in relief,
composed of three distances.
Owing, no doubt, to the difficulty of procuring good binocular
pictures, Mr. Elliot did not see that his contrivance would be very
popular, and therefore carried it no farther. He had never heard of
Mr. Wheatstone’s stereoscope till he saw his paper on Vision
reprinted in the Philosophical Magazine for March 1852, and having
perused it, he was convinced not only that Mr. Wheatstone’s theory
of the instrument was incorrect, but that his claim to the discovery of
the dissimilarity of the images in each eye had no foundation. He
was, therefore, led to communicate to the same journal the fact of
his having himself, thirteen years before, constructed and used a
stereoscope, which was still in his possession. In making this claim,
Mr. Elliot had no intention of depriving Mr. Wheatstone of the credit
which was justly due to him; and as the claim has been publicly
made, we have described the nature of it as a part of scientific
history.
In Mr. Wheatstone’s ingenious paper of 1838, the subject of
binocular vision is treated at considerable length. He gives an
account of the opinions of previous writers, referring repeatedly to
the works of Aguilonius, Gassendi, and Baptista Porta, in the last of
which the views of Galen are given and explained. In citing the
passage which we have already quoted from Leonardo da Vinci, and
inserting the figure which illustrates it, he maintains that Leonardo da
Vinci was not aware “that the object (c in Fig. 2) presented a
different appearance to each eye.” “He failed,” he adds, “to observe
this, and no subsequent writer, to my knowledge, has supplied the
omission. The projection of two obviously dissimilar pictures on the
two retinæ, when a single object is viewed, while the optic axes
converge, must therefore be regarded as a new fact in the theory of
vision.” Now, although Leonardo da Vinci does not state in so many
words that he was aware of the dissimilarity of the two pictures, the
fact is obvious in his own figure, and he was not led by his subject to
state the fact at all. But even if the fact had not stared him in the face
he must have known it from the Optics of Euclid and the writings of
Galen, with which he could not fail to have been well acquainted.
That the dissimilarity of the two pictures is not a new fact we have
already placed beyond a doubt. The fact is expressed in words, and
delineated in drawings, by Aguilonius and Baptista Porta. It was
obviously known to Dr. Smith, Mr. Harris, Dr. Porterfield, and Mr.
Elliot, before it was known to Mr. Wheatstone, and we cannot
understand how he failed to observe it in works which he has so
often quoted, and in which he professes to have searched for it.
This remarkable property of binocular vision being thus clearly
established by preceding writers, and admitted by himself, as the
cause of the vision of solidity or distance, Mr. Wheatstone, as Mr.
Elliot had done before him, thought of an instrument for uniting the
two dissimilar pictures optically, so as to produce the same result
that is obtained by the convergence of the optical axes. Mr. Elliot
thought of doing this by the eyes alone; but Mr. Wheatstone adopted
a much better method of doing it by reflexion. He was thus led to
construct an apparatus, to be afterwards described, consisting of two
plane mirrors, placed at an angle of 90°, to which he gave the name
of stereoscope, anticipating Mr. Elliot both in the construction and
publication of his invention, but not in the general conception of a
stereoscope.
After describing his apparatus, Mr. Wheatstone proceeds to
consider (in a section entitled, “Binocular vision of objects of different
magnitudes”) “what effects will result from presenting similar images,
differing only in magnitude, to analogous parts of the retina.” “For
this purpose,” he says, “two squares or circles, differing obviously
but not extravagantly in size, may be drawn on two separate pieces
of paper, and placed in the stereoscope, so that the reflected image
of each shall be equally distant from the eye by which it is regarded.
It will then be seen that notwithstanding this difference they coalesce
and occasion a single resultant perception.” The fact of coalescence
being supposed to be perfect, the author next seeks to determine the
difference between the length of two lines which the eye can force
into coalescence, or “the limits within which the single appearance
subsists.” He, therefore, unites two images of equal magnitude, by
making one of them visually less from distance, and he states that,
“by this experiment, the single appearance of two images of different
size is demonstrated.” Not satisfied with these erroneous assertions,
he proceeds to give a sort of rule or law for ascertaining the relative
size of the two unequal pictures which the eyes can force into
coincidence. The inequality, he concludes, must not exceed the
difference “between the projections of the same object when seen in
the most oblique position of the eyes (i.e., both turned to the extreme
right or the extreme left) ordinarily employed.” Now, this rule, taken in
the sense in which it is meant, is simply a truism. It merely states
that the difference of the pictures which the eyes can make to
coalesce is equal to the difference of the pictures which the eyes do
make to coalesce in their most oblique position; but though a truism
it is not a truth, first, because no real coincidence ever can take
place, and, secondly, because no apparent coincidence is effected
when the difference of the picture is greater than what is above
stated.
From these principles, which will afterwards be shewn to be
erroneous, Mr. Wheatstone proceeds “to examine why two dissimilar
pictures projected on the two retinæ give rise to the perception of an
object in relief.” “I will not attempt,” he says, “at present to give the
complete solution of this question, which is far from being so easy as
at first glance it may appear to be, and is, indeed, one of great
complexity. I shall, in this case, merely consider the most obvious
explanations which might be offered, and shew their insufficiency to
explain the whole of the phenomena.
“It may be supposed that we see only one point of a field of view
distinctly at the same instant, the one, namely, to which the optic
axes are directed, while all other points are seen so indistinctly that
the mind does not recognise them to be either single or double, and
that the figure is appreciated by successively directing the point of
convergence of the optic axes successively to a sufficient number of
its points to enable us to judge accurately of its form.
“That there is a degree of indistinctness in those parts of the field
of view to which the eyes are not immediately directed, and which
increases with the distance from that point, cannot be doubted; and it
is also true that the objects there obscurely seen are frequently
doubled. In ordinary vision, it may be said, this indistinctness and
duplicity are not attended to, because the eyes shifting continually
from point to point, every part of the object is successively rendered
distinct, and the perception of the object is not the consequence of a
single glance, during which a small part of it only is seen distinctly,
but is formed from a comparison of all the pictures successively
seen, while the eyes were changing from one point of an object to
another.
“All this is in some degree true, but were it entirely so no
appearance of relief should present itself when the eyes remain
intently fixed on one point of a binocular image in the stereoscope.
But in performing the experiment carefully, it will be found, provided
the picture do not extend far beyond the centres of distinct vision,
that the image is still seen single, and in relief, when in this
condition.”[22]
In this passage the author makes a distinction between ordinary
binocular vision, and binocular vision through the stereoscope,
whereas in reality there is none. The theory of both is exactly the
same. The muscles of the two eyes unite the two dissimilar pictures,
and exhibit the solid, in ordinary vision; whereas in stereoscopic
vision the images are united by reflexion or refraction, the eyes in
both cases obtaining the vision of different distances by rapid and
successive convergences of the optical axes. Mr. Wheatstone
notices the degree of indistinctness in the parts of the picture to
which the eyes are not immediately directed; but he does not notice
the “confusion and incongruity” which Aguilonius says ought to exist,
in consequence of some parts of the resulting relievo being seen of
one size by the left eye alone,—other parts of a different size by the
right eye alone, and other parts by both eyes. This confusion,
however, Aguilonius, as we have seen, found not to exist, and he
ascribes it to the influence of a common sense overruling the
operation of physical laws. Erroneous as this explanation is, it is still
better than that of Mr. Wheatstone, which we shall now proceed to
explain.
In order to disprove the theory referred to in the preceding
extract, Mr. Wheatstone describes two experiments, which he says
are equally decisive against it, the first of them only being subject to
rigorous examination. With this view he draws “two lines about two
inches long, and inclined towards each other, on a sheet of paper,
and having caused them to coincide by converging the optic axes to
a point nearer than the paper, he looks intently on the upper end of
the resultant line without allowing the eyes to wander from it for a
moment. The entire line will appear single, and in its proper relief,
&c.... The eyes,” he continues, “sometimes become fatigued, which
causes the line to become double at those parts to which the optic
axes are not fixed, but in such case all appearance of relief
vanishes. The same experiment may be tried with small complex
figures, but the pictures should not extend too far beyond the centre
of the retinæ.”
 Now these experiments, if rightly made and interpreted, are not
decisive against the theory. It is not true that the entire line appears
single when the axes are converged upon the upper end of the
resultant line, and it is not true that the disappearance of the relief
when it does disappear arises from the eye being fatigued. In the
combination of more complex figures, such as two similar rectilineal
figures contained by lines of unequal length, neither the inequalities
nor the entire figure will appear single when the axes are converged
upon any one point of it.
In the different passages which we have quoted from Mr.
Wheatstone’s paper, and in the other parts of it which relate to
binocular vision, he is obviously halting between truth and error,
between theories which he partly believes, and ill-observed facts
which he cannot reconcile with them. According to him, certain truths
“may be supposed” to be true, and other truths may be “in some
degree true,” but “not entirely so;” and thus, as he confesses, the
problem of binocular and stereoscopic vision “is indeed one of great
complexity,” of which “he will not attempt at present to give the
complete solution.” If he had placed a proper reliance on the law of
visible direction which he acknowledges I have established, and
“with which,” he says, “the laws of visible direction for binocular
vision ought to contain nothing inconsistent,” he would have seen the
impossibility of the two eyes uniting two lines of inequal length; and
had he believed in the law of distinct vision he would have seen the
impossibility of the two eyes obtaining single vision of any more than
one point of an object at a time. These laws of vision are as
rigorously true as any other physical laws,—as completely
demonstrated as the law of gravity in Astronomy, or the law of the
Sines in Optics; and the moment we allow them to be tampered with
to obtain an explanation of physical puzzles, we convert science into
legerdemain, and philosophers into conjurors.
Such was the state of our stereoscopic knowledge in 1838, after
the publication of Mr. Wheatstone’s interesting and important paper.
Previous to this I communicated to the British Association at
Newcastle, in August 1838, a paper, in which I established the law of
visible direction already mentioned, which, though it had been
maintained by preceding writers, had been proved by the illustrious
D’Alembert to be incompatible with observation, and the admitted
anatomy of the human eye. At the same meeting Mr. Wheatstone
exhibited his stereoscopic apparatus, which gave rise to an animated
discussion on the theory of the instrument. Dr. Whewell maintained
that the retina, in uniting, or causing to coalesce into a single
resultant impression two lines of different lengths, had the power
either of contracting the longest, or lengthening the shortest, or what
might have been suggested in order to give the retina only half the
trouble, that it contracted the long line as much as it expanded the
short one, and thus caused them to combine with a less exertion of
muscular power! In opposition to these views, I maintained that the
retina, a soft pulpy membrane which the smallest force tears in
pieces, had no such power,—that a hypothesis so gratuitous was not
required, and that the law of visible direction afforded the most
perfect explanation of all the stereoscopic phenomena.
In consequence of this discussion, I was led to repeat my
experiments, and to inquire whether or not the eyes in stereoscopic
vision did actually unite the two lines of different lengths, or of
different apparent magnitudes. I found that they did not, and that no
such union was required to convert by the stereoscope two plane
pictures into the apparent whole from which they were taken as seen
by each eye. These views were made public in the lectures on the
Philosophy of the Senses, which I occasionally delivered in the
College of St. Salvator and St. Leonard, St. Andrews, and the
different stereoscopes which I had invented were also exhibited and
explained.
In examining Dr. Berkeley’s celebrated Theory of Vision, I saw
the vast importance of establishing the law of visible direction, and of
proving by the aid of binocular phenomena, and in opposition to the
opinion of the most distinguished metaphysicians, that we actually
see a third dimension in space, I therefore submitted to the Royal
Society of Edinburgh, in January 1843, a paper On the law of visible
position in single and binocular vision, and on the representation of
solid figures by the union of dissimilar plane pictures on the retina.
More than twelve years have now elapsed since this paper was
read, and neither Mr. Wheatstone nor Dr. Whewell have made any
attempt to defend the views which it refutes.
In continuing my researches, I communicated to the Royal
Society of Edinburgh, in April 1844, a paper On the knowledge of
distance as given by binocular vision, in which I described several
interesting phenomena produced by the union of similar pictures,
such as those which form the patterns of carpets and paper-
hangings. In carrying on these inquiries I found the reflecting
stereoscope of little service, and ill fitted, not only for popular use,
but for the application of the instrument to various useful purposes. I
was thus led to the construction of several new stereoscopes, but
particularly to the Lenticular Stereoscope, now in universal use. They
were constructed in St. Andrews and Dundee, of various materials,
such as wood, tin-plate, brass, and of all sizes, from that now
generally adopted, to a microscopic variety which could be carried in
the pocket. New geometrical drawings were executed for them, and
binocular pictures taken by the sun were lithographed by Mr.
Schenck of Edinburgh. Stereoscopes of the lenticular form were
made by Mr. Loudon, optician, in Dundee, and sent to several of the
nobility in London, and in other places, and an account of these
stereoscopes, and of a binocular camera for taking portraits, and
copying statues, was communicated to the Royal Scottish Society of
Arts, and published in their Transactions.
It had never been proposed to apply the reflecting stereoscope to
portraiture or sculpture, or, indeed, to any useful purpose; but it was
very obvious, after the discovery of the Daguerreotype and
Talbotype, that binocular drawings could be taken with such
accuracy as to exhibit in the stereoscope excellent representations in
relief, both of living persons, buildings, landscape scenery, and every
variety of sculpture. In order to shew its application to the most
interesting of these purposes, Dr. Adamson of St. Andrews, at my
request, executed two binocular portraits of himself, which were
generally circulated and greatly admired. This successful application
of the principle to portraiture was communicated to the public, and
recommended as an art of great domestic interest.