wray, k. b. rethinking scientific specialization. social studies of science, vol. 35, no. 1 (feb.,...

8/18/2019 Wray, K. B. Rethinking Scientific Specialization. Social Studies of Science, Vol. 35, No. 1 (Feb., 2005), Pp. 151-164

1/15

Rethinking Scientific SpecializationAuthor(s): K. Brad WraySource: Social Studies of Science, Vol. 35, No. 1 (Feb., 2005), pp. 151-164Published by: Sage Publications, Ltd.Stable URL: http://www.jstor.org/stable/25046633 .Accessed: 21/06/2013 19:30

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Sage Publications, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to Social Studies of Science.

http://www.jstor.org

This content downloaded from 14 3.106.201.11 on Fri, 21 Jun 20 13 19:30:10 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=sageltdhttp://www.jstor.org/stable/25046633?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/25046633?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=sageltd


2/15

sss

RESEARCH NOTE

ABSTRACT My aim in this paper is to re-examine specialization in science. I arguethat we need to acknowledge the role that conceptual changes can play in the

creation of new specialties. Whereas earlier sociological accounts focus on social and

instrumentalchanges

as the cause of the creation of newspecialties,

I argue that

conceptual changes play an important role in the creation of some scientific

specialties. Specifically, I argue that conceptual developments played an importantrole in the creation of both endocrinology and virology.

Keywords conceptual change, Kuhn, scientific change, scientific specialization,

specialties

Rethinking Scientific SpecializationK. Brad Wray

One of the most striking forms of scientific change is the rapid and

seemingly endless growth of new scientific specialties. Philosophers of

science seldom discuss this dimension of scientific change. Sociologistsand historians, on the other hand, once studied

specializationwith intense

interest. As Harriet Zuckerman (1988) notes, the sociological study of

scientific specialization was itself, for a time, a sociological specialty (1988:

535). Interest in specialization, though, has since passed. This, I believe is

unfortunate.

My aim in this paper is to re-examine specialization in science. In

addition to reviewing the important insights of earlier studies, I want to

emphasize the need to study the role that conceptual changes can play in

the creation of new specialties. Whereas earlier accounts focus on social

changes as the cause of the creation of new specialties, I argue that

conceptual changes have played an important role in the creation of some

scientific specialties.1 The account of the role of conceptual developmentsin the creation of new specialties that I develop and defend here is deeplyindebted to Thomas Kuhn's (2000 [1987], 2000 [1991], 2000 [1993])later writings. Unfortunately, this work has been largely neglected by the

science studies research community.

Social Studies of Science 35/l(February 2005) 151-164? SSS and SAGE Publications (London, Thousand Oaks CA, New Delhi)ISSN 0306-3127 DOI: 10.1177/0306312705045811

www. sagepublications. com


http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/15

152 Social Studies of Science 35/1

Sociological and Historical Studies of Scientific Specialization

In this section I examine a variety of sociological and historical studies ofscientific specialization. I begin by examining two pioneering studies. I

argue that both of these pioneering studies treat the development of a new

specialty as essentially a change in the social organization of science, andthus treat conceptual changes as derivative. Further, both of these studies

imply that there is one cause of the creation of new specialties. I then brieflyreview the sociological studies of specialization that followed these pioneering studies. These second-generation studies challenge the assumption thatthe same type of cause is responsible for the creation of all scientific

specialties. But I argue that, like the pioneering studies, these studiescontinue to treat the conceptual, epistemic, or cognitive changes associated

with scientific specialties as secondary to the social and instrumental

changes associated with the process.Let us begin with the pioneering accounts. According to Joseph Ben

David and Randall Collins (1966 [1991]: 50), the creation of a new

scientific specialty is a consequence of scientists carving out a new professional niche in an effort to create a new social role. When an existing

field shows little promise for career advancement, ambitious and ableyoung scientists will seek means to create a new discipline, field, or

specialty. Ben-David and Collins support their account with a case study of

the creation in the late 1800s of experimental psychology as a discipline in

Germany. As they note, before experimental psychology became a distinct

discipline, 'the subject matter of psychology was divided between speculative philosophy and physiology' (1966 [1991]: 53). In the mid-1800s,

physiology underwent a period of rapid expansion (1966 [1991]: 63). In a

relatively short period of time, between 1850 and 1864, many of theuniversity positions in physiology were filled by young men. As a result,there was little opportunity for career advancement for the upcoming

generation of physiologists (1966 [1991]: 63). Positions were blocked bythe recently appointed generation of men who still had full careers ahead ofthem. This led a number of promising young scientists trained in physiology to create new career opportunities for themselves by applying the

methods of physiology to problems in psychology.Ben-David and Collins suggest that the

developmentof

psychologyis a

typical case of the process by which a new specialty is created. Crowding inan existing field leads young scientists to develop a new specialty in an

effort to secure rewarding employment. Importantly, they insist that con

ceptual developments in the study of the human mind were not responsiblefor the creation of experimental psychology as a discipline (1966 [1991]:50). In fact, Ben-David and Collins believe that, generally, 'the ideas

necessary for the creation of a new discipline are usually available over a

relatively prolonged period of time and in several places' (1966 [1991]:50). They suggest that if ideas were sufficient for the creation of a new

discipline, we should have expected psychology to develop as a disciplinefirst in either France or the UK (1966 [1991]: 67-69).




4/15

Wray: Rethinking Scientific Specialization 153

Let us now consider the second pioneering account of specialization.Derek de Sol?a Price (1986 [1963]) suggests that the chief factor that leads

to the creation of a new specialty is the demand to make effective researchpossible. Scientific research is done by humans, and our limited cognitivecapacities place significant constraints on the organization of science.

Because there is a limit to how much people can read, each scientist can

only attend to a finite and rather small portion of the continuously growing

body of scientific literature. As more and more people get involved in

science, and more and more journals publish more and more papers, eachnew generation of scientists confronts a larger body of scientific literature.

In fact, Price estimates that by the early 1960s there were already more

than 10,000,000 published scientific papers. And the number of publications was doubling every 15 years. Price believes that the various sub-fields

in science are a consequence of scientists carving out manageable bodies of

literature. He hypothesizes that the optimal size for a scientific research

community is between 100 and 200 publishing scientists. Such a commu

nity could keep abreast of the literature they produce (1986 [1963]: 65).There are three features that these pioneering accounts have in com

mon. First, both accounts are premised on the assumption that conceptual

developments in science are not what lead scientists to create new specialties. Hence, both of these pioneering studies privilege social changes in the

creation of a new specialty, and treat conceptual changes as derivative.

Second, both accounts are premised on the assumption that crowding in

an existing field leads to the creation of a new specialty. Price assigns

crowding a different function than that hypothesized by Ben-David and

Collins. Rather than driving young scientists in search of new career

opportunities, Price believes that crowding leads communities of scientists

to narrow their area of research in an effort to prevent themselves frombeing overwhelmed by the growing body of research. Only by narrowingtheir area of research, and thus creating a new specialty, are scientists able

to effectively manage the continuously growing literature. Third, both

accounts are premised on the assumption that there is one type of cause

that leads to the creation of new scientific specialties.The 1970s marked a period of extensive growth in the sociology of

science in general (see Cole & Zuckerman, 1975: 146-48, 165), and in

sociologicalstudies of scientific

specializationin

particular (Zuckerman,1988). The most important of these studies were Nicholas Mullins' (1972)'The Development of a Scientific Specialty', David Edge and Michael

Mulkay's (1976) Astronomy Transformed, Daryl Chubin's (1976) 'The Con

ceptualization of Scientific Specialties', and an anthology edited byLemaine et al. (1976b), Perspectives on the Emergence of Scientific Disciplines.Rather than provide a synthesis of second-generation studies, I want to

draw attention to two ways these studies differed from the pioneeringstudies.

First, a number of second-generation sociological studies examine the

impact that new instruments and the development of instrumentation playin the creation of new specialties. For example, Edge & Mulkay (1976)




5/15


provide a detailed account of the ways in which new instruments shapedradio astronomy. Similarly, John Law (1976) argues that developments in

instrumentation played a significant role in the creation of X-ray crystallography. These studies thus suggest a need to broaden the range of causes ofnew scientific specialties. But they assume that developments in instru

mentation are distinct from conceptual developments (see Law, 1976:

123).

Second, a number of second-generation sociological studies acknowl

edge the complexity of the process by which a new specialty is created

(Lemaine et al., 1976a). In particular, they begin to recognize that the

development of a new specialty involves both social changes, the focus of

the pioneering studies, and conceptual or cognitive changes (Edge &

Mulkay, 1976: 364; Lemaine et al., 1976a: 1;Mulkay & Edge, 1976: 153;

Worboys, 1976: 77). This led some sociologists to question the derivative

role that epistemic developments were assigned in the pioneering studies.

Sociological studies of scientific specialization seemed to privilege the

social dimension of the process to such an extent that it is impossible to

conceive how developments in the content of science could lead to the

creation of a new specialty. As Chubin (1976: 449) explains, sociologicalstudies of scientific specialization take 'social structure as an antecedent of

specialties', which 'seemingly denies the possibility that intellectual events

and the relations they engender give rise to a social structure that we treat

as a specialty.' Becher &Trowler (2001) suggest that Chubin himself was

unable to adequately address this concern.

Unfortunately, the interests and attention of sociologists of science

shifted away from the study of scientific specialization before this concern

was adequately addressed. As Zuckerman (1988: 535) notes, 'with the shift

in research attention to the microsociology of scientific knowledge, thestudy of specialties ... declined markedly amongst sociologists.' I believe

that the study of specialization ended prematurely. Indeed, as Zuckerman

(1988: 535) noted in the late 1980s, 'the reasons that led sociologists of

science to study the development of specialties in the first place still appearto be valid.' In particular, we have yet to develop an adequate under

standing of the role that conceptual changes can play in the creation of new

scientific specialties.

Scientific Specialization and Conceptual Change

In this section, I want to examine an account of the creation of new

specialties that focuses on the epistemic dimension of the change. Thomas

Kuhn discusses specialization in various essays collected together in The

Road Since Structure (2000). Unfortunately, he never wrote a paper de

voted exclusively to the topic. As a result, there has been remarkably little

uptake or response to what he said on the topic. My aim is to synthesize his

remarks into a coherent account of specialization. I believe that Kuhn

provides insight into understanding how conceptual changes can lead to

the creation of new specialties.




6/15


According to Kuhn (2000 [1991]: 97), a new scientific specialty is

sometimes created as a consequence of a scientific revolution. To expressthe

pointin the familiar

thoughoften misunderstood idiom of The Structure

of Scientific Revolutions (Kuhn, 1996 [1962]), when scientists in a field

encounter persistent anomalies and they are unable to resolve the crisis

with the resources provided by the prevailing paradigm, a new paradigmwill inevitably be developed. The new paradigm is designed to resolve the

outstanding anomalies. But sometimes the new paradigm will not be able

to serve the purposes of all those working in the field. As a consequence,

part of the field as it was conceived before the revolution becomes a new

field or specialty. Thus, old paradigms are not necessarily discarded.

Sometimes they come to be employed in a more restricted domain.

As Kuhn developed his account of scientific change in response to

criticism, he modified his view of the nature of scientific revolutions in

subtle but important ways that have important implications for under

standing the process by which a new scientific specialty is created. Let us

begin by examining how he reconceived the notion of a scientific

revolution.

A scientific revolution involves a taxonomic change. For example, in

the Copernican revolution astronomers replaced a lexicon in which

'planet' denotes a satellite of the earth with a lexicon in which 'planet'denotes a satellite of the sun (2000 [1987]: 15). Kuhn believes that 'the

transition to a new lexical structure, to a revised set of kinds, permits the

resolution of problems with which the previous structure was unable to

deal' (2000 [1993]: 250). For example, the new lexicon introduced duringthe Copernican revolution enabled astronomers to explain retrograde

motion. Thus, in Kuhn's developed account of scientific revolutions, the

notion of a taxonomic change replaces the earlier problematic notion of aparadigm change.2

It is worth comparing revolutionary changes with the more common

place discoveries of normal science. In normal science, advances do not

require modifications to existing taxonomies. Rather, as Kuhn (2000

[1987]: 14) explains, discoveries in normal science 'simply [add] to the

way ... antecedently understood variables behave.' Boyle's discovery 'that,

for a given gas sample, the product of pressure and volume was a constant

at constant temperature', is a typical example of a normal non-revolu

tionary scientific discovery.3

According to Kuhn, sometimes the taxonomic changes that are a

consequence of a scientific revolution affect only a subset of a scientific

research community. And, 'what lies outside of [the new taxonomy]becomes the domain of another scientific specialty, a specialty in which an

evolving form of the old kind terms remain in use' (2000 [1993]: 250).

Hence, a new specialty is created when a taxonomic change affects only

part of a research community, and the old taxonomy is still employed

effectively in a restricted domain. Not every revolutionary change permitsthis. For example, Copernicus' proposed taxonomic change could not be

adopted without replacing the pre-Copernican lexicon. In such cases, there




7/15


is no need to create a new specialty, for the whole of the existing research

community adopts the proposed change.4

On this Kuhnian account, specialization involves both epistemic andsocial changes. On the social dimension, scientists who once had regularcontact with each other often have less in common to bring them together.

They may no longer attend the same conferences. They may even beginpublishing in different journals. They may also find communication with

each other more difficult (Kuhn, 1996 [1962]: 177). But, the process of

specialization that Kuhn describes is essentially epistemic in nature. It is an

epistemic shortcoming, the inability of a community of scientists to ade

quately model the domain of their field with the existing taxonomy, whichleads the community to divide and thus create a new specialty. As Kuhn

explains, 'specialization and the narrowing of the range of expertise [are]... the necessary price of increasingly powerful cognitive tools' (2000

[1991]: 98). Confined to the conceptual resources available in the existingscientific taxonomies, the resolution of some outstanding problems may

have been impossible. Certain phenomena may evade our detection unlessa revolutionary change is made to an existing taxonomy. Thus, sometimes

scientists must be prepared to relinquish part of their current domain of

study to those who are willing to employ a new modified taxonomy bettersuited to the study of the recalcitrant phenomena.5

Support for the Kuhnian Account

Kuhn's account of specialization provides us with a means to explain how

conceptual developments in science can contribute to the creation of new

specialties. Though Kuhn does not provide examples from the history of

science tosupport

hisaccount,

I believe there are cases that do in fact

support it. In an effort to defend Kuhn's account, I want to examine twocase studies.

The creation of endocrinology provides a clear illustration of the

process Kuhn describes. As R.A. Gregory (1977: 105) explains:

[T]he discovery in 1902 by Bayliss and Starling ... of the duodenalhormone secretin was ... a signal event in the history of physiology. A

simple experiment ... revealed that the functions of the body were

normallycoordinated not

only bythe nervous

system,but also

bythe

mediation of specific chemical agents formed in, and transmitted from,one organ to others by way of circulation, conveying a message intelligible

only to those cells equipped to capture the 'chemical messenger' and

decipher the encoded instructions for modification of their activity. By the

discovery of what came to be called 'hormones' there was opened a new

era of physiology, the beginning of endocrinology as we know it today.

Before Bayliss and Starling made their discovery, physiologists generallyassumed that the functions of the body were coordinated by the nervous

system, and the taxonomy physiologists worked with reflected this. Conse

quently, in making their discovery, Bayliss and Starling needed to invoke a

new concept, 'hormone' or 'chemical messenger'.




8/15


Bayliss and Starling's discovery significantly altered the field of physi

ology. Part of the field as it was conceived before the discovery was

relinquishedto

scientists preparedto

studythe effects of hormones on the

body. 'Hormone' became a key concept in this new field of study, endocri

nology. But the traditional study of physiology was, to a large degree, left

very much intact as it was before the discovery. Many of the functions of

the body are in fact coordinated by the nervous system, as physiologistsassumed, and physiologists could continue to study various functions of

the body working with that assumption. One significant change that the

discovery brought to physiology was a narrowing of the traditional domain

of the field. The revolutionary discovery of hormones reduced the range of

phenomena that physiologists were expected to explain in terms of the

nervous system. Investigating the various functions of the body that are

coordinated by 'chemical messengers' requires different cognitive tools

than those that were available to physiologists. In fact, the study of these

functions is sufficiently different from that of phenomena that are coordi

nated by the nervous system as to warrant the attention of specialists.Rather than modify the existing taxonomy to accommodate the con

cept 'hormone', physiologists could have tried to explain the observed

phenomena in terms of the then-operative lexicon. Had they tried to dothis, though, it is likely that an accurate understanding would have eluded

them. The existing taxonomy, built on the assumption that the functions of

the body are coordinated by the nervous system, was not fit to explain the

phenomena under consideration.

This example nicely illustrates Kuhn's view. A revolutionary discovery

required significant changes to the taxonomy of a field, with the result that

a new field was born and the domain of the original field was subsequently

truncated. The cognitive tools that were designed for one set of purposes,the study of the nervous system in the functioning of the body, are

discovered to be inadequate for the range of phenomena to which theywere assumed to apply.

There are other episodes in the history of science that follow a similar

path of development, including, for example, the creation of virology as a

separate field of study. In the late 19th century and first half of the 20th

century what we now call viruses were often studied as anomalous phenomena

by bacteriologists, pathologists,and biochemists (Waterson &

Wilkinson, 1978). Some scientists attempted to understand viruses as

microbes. Working with the conceptual and experimental resources of

bacteriology, they were perplexed with the problem of growing viruses in

artificial media. Other scientists attempted to understand viruses as non

organic entities. Working with the conceptual and experimental resources

of chemistry, they were concerned with understanding how a toxic non

living substance could replicate (Hughes, 1977: 79-84, 90). But, in the

1950s 'a fundamental conceptual change occurred with the recognitionthat all viruses share a common structure despite observed differences in

morphology, host specificity and pathological activity' (Hughes, 1977:

100). It was then realized that viruses were a distinct kind of phenomenon,




9/15


and could not be adequately conceptualized either as an organism or as a

toxin. Thus, a new taxonomy had to be constructed to accommodate

'virus'as a

kind.The taxonomy bacteriologists used before the discovery of viruses

remained essentially the same after the discovery. Subsequent research in

bacteriology was thus more or less continuous with the practice before the

discovery of viruses. Bacteriologists, though, were able to relinquish re

sponsibility for explaining certain phenomena that they had previously

regarded as their responsibility, phenomena that had, until then, remained

anomalous. A similar situation occurred in pathology and biochemistry,the other specialties in which scientists investigated the phenomena thatcame to be recognized as viruses.

These two episodes in the history of science illustrate the importantrole that conceptual changes can play in the creation of a new scientific

specialty. Sometimes a new specialty is created as a result of a revolu

tionary discovery. Such a discovery sometimes leads a community of

scientists to split the domain of their field and form two separate research

communities, each pursuing their research with a taxonomy suited to their

needs and interests. The creation of a new specialty is thus one means bywhich scientists are able to develop an understanding of the world. Some

phenomena elude our understanding until scientists make changes to an

existing taxonomy. Such changes allow scientists to approach the study of

recalcitrant phenomena with new conceptual tools. And when a taxonomic

change gives rise to a new specialty, scientists in the parent specialty realize

the limits of their model, and thus relinquish responsibility for explaining

phenomena they are ill-equipped to explain.

Anticipating Criticism

In light of the sociological studies of specialization discussed earlier, I

anticipate two criticisms. Similar criticisms could be raised against either of

the case studies discussed in the previous section, but I will present the

criticisms as they apply to the case of virology. My aim in this section is to

address the criticisms and thus defend the Kuhnian account of specialization that I have developed earlier.

Let us consider the criticisms.First,

onemight argue

that the in

troduction of new instruments and techniques was responsible for the

creation of virology. In support of this view one might cite the fact that the

discovery of viruses depended upon a series of developments made in

filtering, a process that played an integral role in enhancing scientists'

understanding of viruses. One might also cite the fact that the electron

microscope also played an indispensable role in the discovery of viruses

(Hughes, 1977: 96, 98;Waterson &Wilkinson, 1978: 105-06).

Second, one might argue that virology was created, not as the result of

conceptual changes, but rather as a result of crowding in existing fields. In

support of this view, one might cite the fact that even before 1900,Martinus Beijerinck had developed a concept of a non-cellular life form to




10/15


account for 'the agent of tobacco mosaic disease' (Hughes, 1977: 48-51).

Thus, the required conceptual developments seem to have been available

longbefore the creation of

virologyas a

specialtyin the 1950s. Conse

quently, it seems that the creation of virology as a specialty in the 1950s

depended, ultimately, on the migration of physicists into the biologicalsciences. Like the young physiologists who created experimental psychology, these physicists were seeking new career opportunities, and created

them by applying the methods of physics to biological phenomena

(Hughes, 1977: 91).In the remainder of this section, I want to address these criticisms,

beginning with the first. It is undeniable that technological developments

played an important role in the discovery of viruses, but I believe it is a

mistake to regard the technological developments as the ultimate cause of

the creation of virology as a specialty. There are two considerations that

lead me to believe this. First, rather than being an epistemic asset, at times

the reliance on techniques was an impediment to the discovery of viruses.

Before scientists developed the modern concept 'virus', techniquedetermined characteristics, like filterability and microscopic invisibility,were often used to identify substances as viruses. But, as Sally Hughes

(1977: 87) notes, such 'technique-determined physical characteristics ...were inadequate criteria by which to differentiate viruses from all other

types of infectious agents.' That is, these characteristics do not distinguishviruses from other superficially similar but unrelated phenomena. What

scientists lacked was 'information about [the] intrinsic biological properties [of viruses]' (Hughes, 1977: 87). Consequently, I believe it is amistaketo attribute the creation of virology entirely to developments in

instrumentation.

Second, contrary to what is implied by this criticism, I believe that it isa mistake to contrast technological developments with epistemic develop

ments. Earlier sociological studies that emphasized the role that developments in instrumentation played in the creation of new specialties seem to

assume that these are distinct types of developments. This assumption is

most evident in Law's study of X-ray protein crystallography. In fact, Law

(1976: 123) proposes distinguishing between ' technique , theory , and

subject matter specialties', thus implying that the techniques employedin X-ray

crystallographycould have been

developed independentof ac

companying theoretical developments. Contrary to what Law implies, I

believe that developments in instrumentation are often tied to conceptual

developments. That is, developments in instrumentation often either: (1)

depend upon conceptual developments; or (2) occur simultaneous with

conceptual developments. For example, in order to develop the technologyto detect X-rays, R?ntgen had to simultaneously discover X-rays (see

Kuhn, 1996 [1962]: 56-57). Thus, making a sharp distinction between

developments in instrumentation and conceptual developments, as some

of the earlier sociological studies did, ismisleading.6Let us now consider the second criticism. I have three replies to this

criticism. First, I believe that the criticism is based on aWhig reading of




11/15


history. It is only with hindsight, with the knowledge that viruses form a

distinct kind of entity, that we identify all the various investigations that led

to the discovery of viruses as contributions to virology.7 Scientists involvedin these investigations saw things quite differently. Before the 1950s, when

scientists used the term 'virus', it was applied indiscriminately to a varietyof phenomena, many of which have subsequently been identified as bac

teria (Hughes, 1977: 92-95).

Second, though there were earlier anticipations of the modern con

cept, like Beijerinck's proposal which 'agrees to some extent with the

modern definition of the virus' (Hughes, 1977: 57), it is a mistake to saythat he discovered the same thing that was discovered in the 1950s. The

vague understanding that Beijerinck had was, at the time, in conflict with

widely accepted background assumptions, criticized by others, and not

well supported by the then-available data (Hughes, 1977: 57-58; Waterson

& Wilkinson, 1978: 27-30).8Third, many of the scientists who were ultimately responsible for the

creation of virology as a specialty were not looking for new career opportunities. Many of them were well-established researchers, with secure

positions, and actively involved in research in a number of specialties.Hence, the pioneers in virology lacked the incentives to create a new socialniche that had motivated the early founders of experimental psychology.

Consequently, a comprehensive explanation for why virology came to

be created as a new specialty in the 1950s requires reference to both: (1)

conceptual developments in our understanding of viruses; and (2) the

accompanying taxonomic changes required to articulate an adequate un

derstanding of the phenomena.

Concluding Remarks

My analysis emphasizes the epistemic benefits of specialization. It is worth

briefly mentioning the epistemic costs involved. With the creation of a new

scientific specialty the expertise of individual scientists narrows. As a

consequence, scientists are increasingly required to depend upon and trust

the findings of those who work in other specialties.9 This increasing

dependence that characterizes contemporary science is one of the principalcosts of

specializationin science. As new

specialtiesare

created,barriers

are created and these can sometimes be an impediment to the developmentof science. Indeed, some important discoveries have depended uponscientists transcending the barriers created by specialization. For example,

PaulThagard (1999: 84-85, 181) notes that the discovery of the bacterial

theory of ulcers required knowledge from two different medical specialties,

pathology and gastroenterology. Much of the knowledge needed to make

the discovery was available for some time before the discovery was made.

But it was only when two scientists from different specialties worked

collaboratively that the discovery was made.10

In summary, I have developed and defended an account of how

conceptual developments in science can contribute to the creation of a new




12/15


scientific specialty. In developing this account, I have addressed a key

question left unanswered by earlier sociological studies of specialization:

how can conceptual or epistemic developments in science contribute to thedevelopment of a new scientific specialty? I argued that scientific discov

eries will sometimes lead a research community to split into two commu

nities, each subsequently working with a taxonomy suited to its needs and

interests. Further, I argued that this is what led to the creation of both

endocrinology in the early 1900s and virology in the 1950s.

Although my aim has been to account for the role that conceptual

developments can play in the creation of a new specialty, we should be

mindful of an importantinsight

of thesecond-generation sociologicalstudies discussed earlier. The creation of a new specialty is a complex

process, often involving both cognitive and social changes. Even a social

factor like crowding in a field could play an important role in leading to thesorts of conceptual developments that would lead to the creation of a new

specialty. For example, as an existing field gets crowded there may be more

intense competition, which can lead to a significant discovery. But, a

thorough examination of the interaction between social and epistemicfactors in the process of specialization is a topic that goes beyond the scopeof this paper.

Notes

I thank audiences at the following, to whom I presented earlier drafts: the Science and

Humanities Circle at the University of Alberta; the annual meeting of the Canadian Societyfor History and Philosophy of Science; the Philosophy Department Colloquium at the State

University of New York, Oswego; the Pacific Division meeting of the American

Philosophical Association; and the 12th International Congress of Logic, Methodology, and

Philosophy of Science. I also thank the following individuals for their constructive feedback:

Lori Nash, Randall Collins, Kristina Rolin, Paul Teller, David Hull, Lisa Gannett, Leslie

Cormack, Robert Smith, Patrick McGivern, Robert Wilson, and Alex Reuger. In addition, I

thank the two anonymous referees for Social Studies of Science who provided insightfulcomments that improved the paper. Harold Kincaid and David Vampola also provided

insight and encouragement as I worked on the project. Finally, I thank the Dean of the

College of Arts and Sciences and the Office of International Studies at SUNY-Oswego for

their financial support to cover travel costs to two of the conferences listed earlier.

1. The important distinction here is not between 'externar and 'internal' factors. Indeed,as Hull et al. (1978) argue, it is often difficult to determine whether or not a factor is

to count as external. For example, scientists have frequently been influenced by'extrascientific beliefs', including metaphysical principles, and socioeconomic and

religious beliefs (1978: 717). From our perspective today, what clearly looks like an

external factor was often regarded as internal by those influenced by it. The distinction

that ismy concern here is between social factors and epistemic factors.

2. Kuhn borrowed the term 'paradigm' from 'the standard examples employed in

language teaching' (1977a: xix; 1996 [1962]: 23). But, as he explains, 'paradigms took

on a life of their own .... Having begun simply as exemplary problem solutions, they

expanded their empire to include, first, the classical books in which these accepted

examples initially appeared and, finally, the entire global set of commitments shared by

the members of a particular scientific community' (1977a: xix). Steven Weinberg(1998) notes that James Conant had raised concerns about Kuhn's use of the term

paradigm even before the publication of Structure. Paradigms still have an importantrole to play in Kuhn's final statement on scientific change. As widely accepted




13/15


exemplars that guide a community of scientists in solving unresolved problems, they are

the locus of consensus. It is in this respect that Kuhn's view differs most from the

positivists. The positivists assumed that scientists were united by either an agreed upon

method, or a set of sentences that constitute the accepted theory.

3. Kuhn's amended account of scientific revolutions alleviates some of his critics' worries.

For example, criticizing Kuhn's earlier paradigm-related account of scientific

revolutions, Edge & Mulkay (1976: 392) argue that 'it is far from easy to state

unambiguously what kinds of changes should, from Kuhn's perspective, entail the

occurrence of a revolution.' Given Kuhn's developed account of scientific revolutions,

all and only those changes that require taxonomic changes will count as revolutionary.4. Mulkay (1975: 512) describes Kuhn's paradigm-related account of scientific

development as a closed model, for it stresses 'the existence of scientific orthodoxies.'

Mulkay (1975: 513) believes that such accounts are ultimately unacceptable for they'make scientific innovation highly problematic'. Instead he recommends a branching

model of scientific development, according to which 'new problem areas are regularlycreated and associated social networks formed' (1975: 520). The Kuhnian account of

specialization presented here is more aptly described as a branching model.

5. Abbott (2001) has recently developed an account of the dynamics of scientific fields

that attributes an important role to conceptual changes. But unlike the Kuhnian

account presented here, Abbott's account applies to social sciences only, and it is

principally concerned with the dynamics within a field, rather than the dynamics that

lead to the creation of new fields.

6. I thank one of the referees for Social Studies of Science for advice on developing a

response to this first criticism, and in particular for suggesting the second response.

7. Brannigan (1981) makes a similar point with respect to Mendel's alleged contributions

to genetics.8. As Waterson & Wilkinson (1978: 27) note, a number of Beijerinck's contemporaries

wh^were researching the tobacco mosaic disease, including C.J. Koning, K.G.E.

Heintzel, Friedrich Hunger, and Adolf Mayer, did not even understand 'Beijerinck's

flight of thought'. Others, including Emile Roux, Dmitri Ivanovski, and Eugenio

Centanni, were critical of Beijerinck's conceptual innovation (Waterson & Wilkinson,

1978: 27-28).9. On the role of trust in science, see Steven Shapin's (1994) A Social History of Truth.

10. Thagard believes that it is unlikely that either scientist would have made the discovery

working alone. Recognizing the predicament that is created by specialization, Swanson

& Smalheiser (1997) have developed computer software to identify connections

between bodies of scientific literature from different fields that might lead to scientific

discoveries.

References

Abbott, Andrew (2001) Chaos of Disciplines (Chicago, IL: University of Chicago Press).

Becher, Tony & Paul R. Trowler (2001) Academic Tribes and Territories (Buckingham, Bucks.:The Society for Research Into Higher Education and Open University Press).

Ben-David, Joseph & Randall Collins (1966 [1991]) 'Social Factors in the Origins of a New

Science: The Case of Psychology', in J. Ben-David (ed.), Scientific Growth: Essays on

the Social Organization and Ethos of Science (Berkeley, CA: University of California

Press): 49-70.

Brannigan, Augustine (1981) The Social Basis of Scientific Discovery (Cambridge: Cambridge

University Press).

Chubin, Daryl E. (1976) 'The Conceptualization of Scientific Specialties', The Sociological

Quarterly 17: 448-76.

Cole, Jonathan R. & Harriet Zuckerman (1975) 'The Emergence of a Scientific Specialty:The Self-Exemplifying Case of the Sociology of Science', in L.A. Coser (ed.), The Idea

of Social Structure: Papers inHonor of Robert K Merton (NewYork: Harcourt Brace

Jovanovich): 139-74.




14/15


Edge, David O. & Michael J. Mulkay (1976) Astronomy Transformed: The Emergence of Radio

Astronomy in Britain (New York: John Wiley and Sons).

Gregory, R.A. (1977) 'The Gastrointestinal Hormones: A Historical Review', in A.L.

Hodgkin, A.F. Huxley, W Feldberg, W.A.H. Rushton, R.A. Gregory & R.A. McCance

(eds). The Pursuit of Nature: Informal Essays on the History of Physiology (Cambridge:

Cambridge University Press): 105-32.

Hughes, Sally Smith (1977) The Virus: A History of the Concept (New York: Science History

Publications).

Hull, David L., Peter D. Tessner & Arthur M. Diamond (1978) 'Planck's Principle', Science

202 (17 November): 717-23.

Kuhn, Thomas S. (1977a) 'Preface', inT.S. Kuhn, The Essential Tension: Selected Studies in

Scientific Tradition and Change (Chicago, IL: University of Chicago Press): ix-xxiii.

Kuhn, Thomas S. (1977b) The Essential Tension: Selected Studies inScientific

Tradition and

Change (Chicago, IL: University of Chicago Press).

Kuhn, Thomas S. (1996 [1962]) The Structure of Scientific Revolutions, 3rd edn (Chicago,IL: University of Chicago Press).

Kuhn,Thomas, S. (2000 [1987]) 'What are Scientific Revolutions?', inT.S. Kuhn, The

Road Since Structure: Philosophical Essays, 1970-1993, with an Autobiographical

Interview, edited by J. Conant & J. Haugeland (Chicago, IL: University of Chicago

Press): 13-32.

Kuhn,Thomas S. (2000 [1991]) 'The Road Since Structure1, inT.S. Kuhn, The Road Since

Structure: Philosophical Essays, 1970-1993, with an Autobiographical Interview, edited by

J.Conant &

J. Haugeland (Chicago,IL:

Universityof

Chicago Press): 90-104.Kuhn, Thomas S. (2000 [1993]) 'Afterwords', inT.S. Kuhn, The Road Since Structure:

Philosophical Essays, 1970-1993, with an Autobiographical Interview, edited by J. Conant

& J. Haugeland (Chicago, IL: University of Chicago Press): 224-52.

Kuhn, Thomas S. (2000) The Road Since Structure: Philosophical Essays, 1970-1993, with an

Autobiographical Interview, edited by J. Conant & J. Haugeland (Chicago, IL:

University of Chicago Press).

Law, John (1976) 'The Development of Specialties in Science: The Case of X-Ray Protein

Crystallography', in Gerard Lemaine, Roy MacLeod, Michael Mulkay & Peter

Weingart (eds), Perspectives on the Emergence of Scientific Disciplines (Chicago, IL: Aldine

Publishing Company): 123-52.Lemaine, Gerard, Roy MacLeod, Michael Mulkay & Peter Weingart (1976a) 'Problems in

the Emergence of New Disciplines', in Gerard Lemaine, Roy MacLeod, Michael

Mulkay & Peter Weingart (eds), Perspectives on the Emergence of Scientific Disciplines

(Chicago, IL: Aldine Publishing Company): 1-23.

Lemaine, Gerard, Roy MacLeod, Michael Mulkay & Peter Weingart (eds) (1976b)

Perspectives on the Emergence of Scientific Disciplines (Chicago, IL: Aldine Publishing

Company).

Mulkay, M.J. (1975) 'Three Models of Scientific Development', Sociological Review 23:

509-26.

Mulkay, M.J. & D.O. Edge (1976) 'Cognitive, Technical and Social Factors in the Growthof Radio Astronomy', in Gerard Lemaine, Roy MacLeod, Michael Mulkay & Peter

Weingart (eds), Perspectives on the Emergence of Scientific Disciplines (Chicago, IL: Aldine

Publishing Company): 153-86.

Mullins, Nicholas C. (1972) 'The Development of a Scientific Specialty: The Phage Groupand the Origins of Molecular Biology', Minerva: Review of Science, Learning and Policy

X: 1, 51-82.

Price, Derek de Sol?a (1986 [1963]) Little Science, Big Science ... and Beyond (New York:

Columbia University Press).

Shapin, Steven (1994) A Social History of Truth: Civility and Science in Seventeenth Century

England (Chicago, IL: University of Chicago Press).

Swanson, Don R. & Neil R. Smalheiser (1997) 'An Interactive System for Finding

Complementary Literatures: A Stimulus to Scientific Discovery', Artificial Intelligence91: 183-203.




15/15


Thagard, Paul (1999) How Scientists Explain Disease (Princeton, NJ: Princeton University

Press).

Waterson, A.P. & L. Wilkinson (1978) An Introduction to the History of Virology (Cambridge:

Cambridge University Press).

Weinberg, Steven (1998) 'The Revolution that Didn't Happen', The NewYork Review ofBooks XLV (8 October): 15, 48-52.

Worboys, Michael (1976) 'The Emergence of Tropical Medicine: A Study in the

Establishment of a Scientific Specialty', in Gerard Lemaine, Roy MacLeod, Michael

Mulkay & Peter Weingart (eds), Perspectives on the Emergence of Scientific Disciplines

(Chicago, IL: Aldine Publishing Company): 75-98.

Zuckerman, Harriet (1988) 'The Sociology of Science', in N.J. Smelser (ed.), The Handbook

of Sociology (London: SAGE Publications): 511-74.

K. Brad Wray isAssistant Professor in the Department of Philosophy at theState University of New York, Oswego. His research addresses the followingissues: the interaction between epistemic and social factors in science; therole of collaborative research in science; invisible hand explanations for thesuccess of science; the relationship between age and scientific productivity;and the viability of functional explanations of social phenomena.

Address: Department of Philosophy, State University of New York, Oswego,128 Piez Hall, Oswego, NY 13126, USA; email: [email protected]

wray, k. b. rethinking scientific specialization. social studies of science, vol. 35, no. 1 (feb.,...

Documents