Automobiles

12 pages
8 views

Credit for discoveries: Citation data as a Basis for History of Science Analysis

of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Credit for discoveries: Citation data as a Basis for History of Science Analysis
Transcript
  B.I.B. LINDAHL, AANT ELZINGA and ALFRED WELLJAMS-DOROF CREDIT FOR DISCOVERIES: CITATION DATA AS A BASIS FORHISTORY OF SCIENCE ANALYSIS ABSTRACT. Citation data have become an increasingly significant source of informationfor historians, sociologists, and other researchers studying the evolution of science. In thepast few decades elaborate methodologies have been developed for the use of citation datain the study of the modern history of science. This article focuses on how citation indexesmake it possible to trace the background and development of discoveries as well as toassess the credit that publishing scientists assign to particular discoverers. Kuhn’s notionof discovery is discussed. The priority dispute over the discovery of the AIDS virus is usedas an example.KEY WORDS: AIDS, citation analysis, history of science, scientific discovery, NobelPrize 1. INTRODUCTIONIn an article in  Science , published shortly prior to his celebrated work  The Structure of Scientific Revolutions , Thomas Kuhn calls attention to thedifficulty of determining the influence of individual scientists on particulardiscoveries in the history of science: “To the historian discovery is seldoma unit event attributable to some particular man, time, and place”. 1 Discov-ering something, Kuhn argues, is not a single simple act, but a processextending over time and often involving a number of people, whose indi-vidual contributions may not always be easily defined. 2 In addition, asRobert Merton points out, “the history of science record[s] thousands of instances of similar discoveries having been made by scientists workingindependently of one another”. 3 Against this background, it may seemdifficult, if not impossible, to assign credit to an individual scientist for aparticular discovery. This problem has only increased, with the tremendousexpansion of science and collaborative research in the twentieth century.Nevertheless, discoveries are often attributed to individual scientists byhistorians, contemporary scientists, and awards committees, and prioritydisputes are not uncommon.In the study of modern history of science,  citation indexes , recordingthe reference connections between publications, provide a powerful tool Theoretical Medicine and Bioethics  19:  609–620, 1998.© 1998  Kluwer Academic Publishers. Printed in the Netherlands.  610  B.I.B. LINDAHL ET AL. for analyzing to what extent a certain scientist has contributed to a partic-ular discovery. We focus here on the problem of using citation indexesto trace the background and development of discoveries and to assessthe credit publishing scientists assign to particular discoverers. We brieflydiscuss the very notion of scientific discovery, and compare Kuhn’s notionwith a narrower concept that is commonly employed in science. Using thepriority dispute over the AIDS virus discovery as an example, we discusshow discoveries can be traced, and recognition of particular discoverersassessed, with the use of citation data.2. THE NOTION OF SCIENTIFIC DISCOVERYKuhn illustrates the problem of determining the influence of individualscientists on a particular discovery with the discovery of oxygen, to whichat least three scientists can be said to have a legitimate claim, Carl Scheele,Joseph Priestley, and Antoine Lavoisier: The earliest of the claimants to prepare a relatively pure sample of the gas was the Swedishapothecary, C. W. Scheele. We may, however, ignore his work since it was not publisheduntil oxygen’s discovery had repeatedly been announced elsewhere [...] Priestley’s claimto the discovery of oxygen is based upon his priority in isolating a gas that was later recog-nized as a distinct species. But Priestley’s sample was not pure [...] Besides, if Priestleywas the discoverer, when was the discovery made? In 1774 he thought he had obtainednitrous oxide, a species he already knew; in 1775 he saw the gas as dephlogisticated air,which is still not oxygen or even, for phlogistic chemists, a quite unexpected sort of gas.Lavoisier’s claim may be stronger, but it presents the same problems. If we refuse the palmto Priestley, we cannot award it to Lavoisier for the work of 1775 which led him to identifythe gas as the “air itself entire.” Presumably we wait for the work of 1776 and 1777 whichled Lavoisier to see not merely the gas but what the gas was. Yet even this award couldbe questioned, for in 1777 and to the end of his life Lavoisier insisted that oxygen was anatomic “principle of acidity” and that oxygen gas was formed only when that “principle”united with caloric, the matter of heat. Shall we therefore say that oxygen had not yet beendiscovered in 1777? 4 As appears from this example, Kuhn assumes that the issue of prioritydepends on scientists’ interpretation of the findings. In order to have madea discovery, a scientist (or, we may add, a team of scientists) ought to haverecognized “both  that   something is and  what   it is”. 5 According to Kuhn’sconcept of discovery, the scientist (or team) would even need to have beenamong the first (if not  the  first) to detect the anomaly that initiated theresearch: Discovery commences with the awareness of anomaly, i.e., with the recognition that naturehas somehow violated the paradigm-induced expectations that govern normal science. Itthen continues with a more or less extended exploration of the area of anomaly. And it  CREDIT FOR DISCOVERIES  611 closes only when the paradigm theory has been adjusted so that the anomalous has becomethe expected. 6 Kuhn’s notion of discovery seems to include: (i) the srcinal confronta-tion with the problem that occasioned the research; (ii) the suggestion of solutions (hypotheses) to the problem; (iii) the testing of the suggestedsolutions/hypotheses; and (iv) the adjustment (or replacement) of the“paradigm theory” that contributed to the occurrence of the problem inpoint (i).Given this inclusive concept of discovery it may indeed be quite diffi-cult to determine to what extent a certain scientist has contributed to aparticular discovery. The problem of defining the notion of discovery isbrought to a head in deliberations and decisions of awarding scientificprizes. A clear example is the Nobel Prize. Kuhn’s inclusive notion of discovery, and his thesis about the difficulty of determining the influenceof individual scientists on particular discoveries, is of particular interestin this connection. For more than 90 years, the Nobel Committees haveawarded individual scientists for particular discoveries. 7 As appears fromFranz Luttenberger’s 8 analysis of the criteria for selecting the winners of the Nobel Prize in Physiology or Medicine in 1908, in awarding PaulEhrlich the Nobel Committee seems to have taken into account a process atleast as comprehensive as the one Kuhn calls “discovery”. A closer exam-ination of the Nobel Committees’ deliberations may, therefore, provide abasis for challenging Kuhn’s thesis. 9 From this point of view, it wouldbe particularly interesting to study the differences between shared andunshared prizes. 10 Kuhn’s comprehensive notion of discovery does not seem to be gener-ally employed in science. The discovery seems often to be restricted to a test result  , e.g., in the case of oxygen, the mere isolation of the gas andonly the earliest speculations as to its nature (which may later turn outto be wrong and which may not influence the “paradigm theory”). Thesrcinal observations that gave rise to the problem investigated, and thesubsequent theory change leading up to the most recent understanding of the phenomenon, seldom, if ever, enters this concept of discovery.Take for example the controversy over the discovery of the AIDSvirus. The dispute concerned who was the first  to isolate the virus , RobertGallo’s group at the National Cancer Institute in Bethesda, Maryland, orLuc Montagnier’s group at the Pasteur Institute in Paris. 11 The recog-nition of the clinical syndrome of AIDS in 1981, 12 the cause of whichdefied explanation (a true anomaly, in Kuhn’s sense), and the disclosurein 1985 of the genetic sequence of the respective viruses (i.e. the determi-nation of   what   kind of virus they had isolated), 13 were not at issue in the  612  B.I.B. LINDAHL ET AL. controversy. These achievements were not seen as parts of the discoveryper se.Although the number of scientists and events can be reduced by apply-ing the more narrow notion of discovery, confined to a test result, there willstill be a great number of scientists whose influence on the discovery has tobe assessed. In the discovery of the AIDS virus, even after the time of thediscovery has been determined to a specific year – officially 1983, whenthe results were published – both the problem of independent discoveriesand the problem of discoveries being made by research teams remain to bedealt with.Kuhn seems to agree with the prevalent practice in science to let thetime of publication of a discovery decide the issue of priority. As we notedearlier, regarding the discovery of oxygen, Kuhn suggests that we ignoreScheele’s work “since it was not published until oxygen’s discovery hadrepeatedly been announced elsewhere”. 14 When it comes to discussing the question of priority for the discoveryof the AIDS virus, we will focus on the results published simultaneouslyin 1983. However, in a recent study of the allocation of credits for thisdiscovery, 15 the number of citations to a paper published in 1983 16 iscompared with the number of citations to a paper published in 1984. 17 Rawling refers to a previous study of AIDS research, 18 that, accordingto Rawling, “have labeled this article 19 as the discovery article of Gallo’steam”. 20 We do not agree with Rawling on this point. Although the paper byGallo et al. in 1984 21 more strongly indicates that the virus discoveredcauses AIDS, the isolation of the virus was published already in a paperby Gallo et al. in 1983. 22 To us, it is also clear that Small and Greenlee 23 use Gallo et al. (1983), and not the 1984 paper, as the discovery articleof this team. Small and Greenlee refers to Gallo et al. (1983) as oneof “[t]he three 1983 papers that embody the retroviral hypothesis forAIDS ...”. 24 3. THE USE OF CITATION INDEXES3.1.  Tracing Discoveries Citation counting has been used for more than half a century. 25 The ideato create a citation index, that could be used for historical research, wasdiscussed already in 1955. 26 However, not until 1964 was the requiredmethodology for tracing discoveries, by the citation network technique,  CREDIT FOR DISCOVERIES  613developed and practically tested, 27 and the use of computers for this task was first conceived in 1967. 28 The basic idea of this approach is that discoveries are traced throughthe impact they have had on scientific development. In practice this meansthat the srcinal publications are identified through the clusters of papersciting them. A cluster emerges when two or more papers are cited together(co-cited) by several papers. Each cluster indicates that papers have asubject in common, e.g. a recent discovery. The srcinal publications of thediscovery will stand out as the most frequently co-cited papers; they willconstitute the core literature of the cluster. 29 The chronology and interde-pendence of such clusters have been mapped for entire research fields,for example for DNA research 1820–1962, 30 opiate receptor research1974–77, 31 plant protoplasm research 1973–80, 32 and AIDS research1982–88. 33 Based on the combined  Science Citation Index  and  Social SciencesCitation Index , the AIDS research study 34 showed a growth in the numberof clusters from only one, with 14 core documents, in 1982, to 80, and669 core documents, in 1988. Already in 1983 eight clusters and 80 coredocuments could be discerned.Two clusters dominated the picture in 1983. One was a further develop-ment of the cluster from 1982, focusing on the clinical syndrome of AIDS.The other was new and focused on etiology. The three most frequentlyco-cited papers in this cluster suggested that AIDS was caused by a retro-virus: Essex et al. (1983), 35 Gallo et al. (1983) 36 and Barré-Sinoussi et al.(1983). 37 3.2.  Assessing Credit for Discoveries In 1986 Myron Essex, Robert C. Gallo, and Luc Montagnier shared theAlbert Lasker Clinical Medical Research Award for their research on theAIDS virus. 38 They had all contributed to the discovery of the AIDS virus.Their respective roles in making the discovery had been different, however.This appears from the motivations of their awards, and has also beencommented upon in previous accounts of the discovery. 39 Interestingly, according to the citations received from 1983 through1994 in the combined  Science Citation Index  and  Social Sciences Cita-tion Index , there are also great differences in the scientific community’srecognition of the three core papers of the discovery (Table I).Clearly, the scientific community preferentially recognised the paperfrom Montagnier’s group (the Barré-Sinoussi paper 40 ) as indicated bycitation counts. As appears from Rawling’s study 41 of the AIDS viruscontroversy, from 1986 through 1992 the citations to Barré-Sinoussi et
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x