Citations and Scientific Indexing

A classic analytical distinction between a citation and a reference reads: ”if paper R contains a bibliographic footnote using and describing paper C, then R contains a reference to C, and C has a citation from R.” According to this, citation and referencing are relations among published texts. But whereas referencing is an intratextual relation between a written word, statement, description, or even an entire argument and a bibliographic reference, citation is an extratextual relation between a complete piece of scientific literature, namely a book or a journal article (just to mention classic forms), and many other pieces of literature of a latter publication date. A reader can easily see a reference by inspecting a text; after all, referencing is a technical standard for editing publish able texts. But a reader cannot see citation directly. Whereas referencing, when it happens, takes place within a singular piece of edited and published material, citation, when it happens, is something that takes place across a section of published literature and it becomes visible for a reader, so to speak, as long as some kind of bibliographic control of that literature can be exerted. This is the purpose that a citation index accomplishes. A citation index is a form to organize and display a body of bibliographical references. These references are collected from reference lists of journal articles, books, and so on, and then organized alphabetically by author. Once this list is ready, under each of the entries one finds the record of published works that have cited them at some point.

A citation index is one of several tools developed throughout centuries of printing for con trolling the literature bibliographically. In fact, among the first textual devices functioning as indexes, the Roman Catholic Index Librorum Prohibitorum (Rome, 1559) constitutes a good example of the double sided nature of this textual technology that is at once social and textual. Journals, too, since their very inception in seventeenth century Europe were used as bibliographical control tools for a growing mass of published material until a whole range of secondary serials (abstracting and indexing services) was launched toward the end of the nineteenth century. However, the bibliographical  control  of scientific literature through citation   indexing   was   a   twentieth century achievement in which developments in both journal editorial standards (Bazerman 1984) and computing technology concurred in the making of scientific indexing through citation an empirical possibility (Garfield 1955). In contrast with the Roman Index, which served the purpose of catching what the Catholic Church considered as heretical texts, authors, and distribution networks during an age of religious turmoil, an index to scientific literature serves to identify significant scientific contributions, significant   scientific   authors,   and relevant sociocognitive  and   sociotechnical networks. And it is because of this that when a citation index for science became a technical reality and a prosperous commercial enterprise in the realm of information retrieval by 1964 (Garfield 1979), further implications were immediately sought for the sociology of science and for science policymaking and management, where the notion of citation and its associated technical procedure of indexing have been consequential.

As an information retrieval tool, citation indexing of scientific literature has empowered scientists by providing them with means and search criteria for taking hold over an increasingly growing mass of scientific literature, and thus has become instrumental for scientific research. As a methodological operationalization within sociology it contributed substantially to the advancement of the empirical investigations on the normative structure of science. And yet, further considerations on the notion of referencing as a rhetorical resource (Gilbert 1977; Latour & Fabbri 1977; Woolgar 1988), on the one hand, and citation as part of the credibility cycle in science (Latour & Woolgar 1979), on the other, opened up new theoretical avenues for exploring sciences as social phenomena. As for science policy and management concerns, citation indexing of scientific literature, used as an evaluative tool often leading to research resource allocation, has proved to be a widespread practice and certainly an effective mechanism of social control, whether one likes it or not. This revolutionary technique for indexing scientific literature pushed forward quantitatively oriented studies on the history of science as well. Indeed, in conjunction with several other notions such as scientific productivity and scientific growth, citation (and co citation) analysis allowed the construction of the scientific literature itself as a knowledge object deserving systematic investigation by scientometrics.

For the main purposes concerning us here, some sociologists used citation and scientific indexing at first as a means for developing analytic methods, based on observable patterns traceable in the literature, for studying the actual operation of norms and values responsible for the emergence of science as a social institution in modern societies, and therefore deserving of sociological examination. The interpretation of citation as an expression of an institutionalized pattern of conduct (acknowledging the sources on which one’s work has been built) in science was set down by Merton and Zuckerman in a classic paper on the issue (Zuckerman & Merton 1973) and presented in context later in Merton’s episodic memoirs (Merton 1977). When a scientific author references someone else’s work in his or her own paper, this author is at least acknowledging authorship (a property right) to someone else. As is immediately obvious, in order to be granted with intellectual property rights over a piece of literature, a scientist must become an author in the first place; science is published knowledge. Thus, Zuckerman and Merton conceived of publication in science as an ingenious procedure for socially granting intellectual property in science, and at the same time contributing to advancing certified knowledge by making it public. If publication grants a basis for claiming intellectual right, it is through citation that this right is socially enjoyed, though not being cited is like being the owner of a useless result from a cognitive point of view. The idea that the more a paper or book is cited, the more impact it has had within a field of studies, and the greater its influence in the community, led some sociologists to conclude that social standing and mobility within the scientific community depend, to a great extent, on the quality of a scientist’s work, as this quality can be ascertained objectively  through  citation  counting.   On this ground, the use of citation counting as an evaluative tool among science policymakers and science managers became widespread. However, it is important to notice that normative sociologists’ interest in scientific quality was related to a more far reaching research agenda on the institutionalization of science. Functionalists used citation measurements not for the sake of measuring and ranking people, institutions, and countries but as empirical evidence supporting an explanation of social stratification in science as a result of an operating structure of institutionalized norms and internalized values.

The idea of citation counting as an objective measurement of scientific quality as well as the attempts at writing a ”scientific history” of science, or drawing maps of knowledge using citation and co citation analysis, has been as controversial as it has been fruitful (Edge 1977). The citation debate opened new avenues for studying science in which emphasis is placed on the mirror image of citation: referencing. Normative uses of citations assume that cognitive and technical standards for research performance and for evaluating scientific results are shared by participants. However, when focusing on scientific practice as it actually takes place in laboratory settings, some sociologists found that those standards were outcomes of social negotiation among participants and therefore context dependent (Mulkay 1991 [1976]). In this light, semiotic minded analyses of scientific texts were undertaken to substantiate the view according to which scientists’ claims of objectivity with regard to the facts presented in their published papers are actually constructed in the text, and referencing is one among several other ”stylistic” resources for doing so (Woolgar 1988). The study of scientific writing showed that the use of references might be understood as a rhetorical resource used in scientific papers whose aim is to persuade readers on different matters. These studies have shown a consistent ”style” in scientific writing that starts by portraying a reported result as a genuine novelty. This is often achieved by reviewing the current state of the art in the introductory section of a paper where referencing is used profusely. In the material and methods section of the paper, referencing serves also the purpose of stating that adequate and authoritative techniques were employed. Very often, too, scientific authors typically show, usually in a concluding section, how their findings illuminate or solve problems reported in current literature, also referenced in the paper, as a means to substantiate the importance of the new published result (Woolgar 1988). A paper’s reference list, then, provides rhetorical force for its arguments by appealing to a persuasive community made out of references that partially set the context of reading for the audience. Thus, by using references, scientists manage to assemble a network which is at once social and technical, a network that may be adequately deployed to support the facticity of a particular statement, or to deny or undermine the facticity of someone else’s statement. Thus the intended audience of a paper is made up of those who are collectively of the opinion that the referenced papers on the list deserve a citation (Gilbert 1977), and those who have been persuaded of this or who find it useful for the advancement of their particular claims (Latour & Woolgar 1979). The more citations a paper receives over time has nothing to do with its objective quality. Citation counting provides only secondary evidence of the success of a particular scientist, research team, or laboratory in advancing their interests which, in the last analysis, can be reduced to remaining well positioned within a continuous cycle of credibility gaining as a means for actually being able to do more science and starting the cycle once more.

Almost without exception, studies concerned with one of the several varieties of citation analysis have been of an empirical nature and based on the counting of the number of citations. However, despite the several warnings that the citation debate arose on the inadvisability of employing citation data without a sound  theoretical  underpinning,  little progress has been made toward the formulation of a ”theory of citation.”  Lately, though (Leydesdorff 1998), a reflexive view on citation states that the quest for a theory of citation presumes that citations themselves should be explained. The moment one starts to count citations to a published work, one is assuming that this tally tells us something about the cited text, about its position in a host of networks: semantic networks, networks of journals, institutional networks (Wouters 1998). If citation analysis is just a tool for explaining, for example, the growth of science, a ”theory of citation” cannot be more than a methodological reflection designed to improve the accuracy of this measurement, merely a technical issue (Woolgar 1991). But when one raises questions such as whether citations indicate ”impact,” ”influence,” or ”quality,” one is in need of a clear definition of these concepts with reference to units of analysis. The reflexive lesson to be learned from the citation debate is that the functions of citations are expected to be different when different contexts or different levels of aggregation are studied, as suggested in the above competing sociological interpretations of citation. Citation analysis is based on a theoretical reflection of scientific practices that have been shaped historically, but the historical, philosophical, and/or sociological positions taken by citation analysts, however, have usually remained implicit. Understanding citation in terms of interacting networks of authors and texts over time enables the possibility of a new theory of citation as a ”dynamic operation that allows for reduction of complexity in various contexts at the same time. The dynamic perspective of selections operating upon selections in other networks accounts for the character of citations as statistical indicators, for their specificity and for their multicontextuality” (Leydesdorff 1998).


  1. Bazerman, C. (1984). Modern Evolution of Experimental Report in Physics: Spectroscopy Articles in Physical Review, 1893-1980. Social Studies of Science 14: 163-96.
  2. Edge, D. (1977) Why I Am Not a Co-Citationist. In: Essays of an Information Scientist (1977 1978), Vol. 3. ISI Press, Philadelphia, pp. 240-6.
  3. Garfield, E. (1955) Citation Index for Science. Science 122: 108-11.
  4. Garfield, E. (1979) Citation Index: Its Theory and Application in Science, Technology, and Humanities. Wiley, New York.
  5. Garfield, E., Merton, R. K., Malin, M. V., & Small, H. G. (1978) Citation Data as Science Indicators. In: Elkana, Y., Lederberg, J., Merton, R. K., Thackray, A., & Zuckerman, H. A. (Eds.), Toward a Metric of Science: The Advent of Science Indicators. Wiley, New York, pp. 179-207.
  6. Gilbert, G. N. (1977) Referencing as Persuasion. Social Studies of Science 7: 113-22.
  7. Latour, B. & Fabbri, P. (1977) La rhetorique de la science: pouvoir et devoir dans un article de la science exacte. Actes de la Recherche en Sciences Sociales 13: 81-95.
  8. Latour, B. & Woolgar, S. W. (1979) Laboratory Life: The Construction of Scientific Facts. Sage, Beverly Hills.
  9. Leydesdorff, L. (1998) Theories of Citation? Scientometrics 43: 5-25.
  10. Merton, R. K. (1977) The Sociology of Science: An Episodic Memoir. In: Merton, R. K. & Gaston, J. (Eds.), The Sociology of Science in Europe. Southern Illinois University Press, Feffer and Simons, Carbondale and London, pp. 3-141.
  11. Mulkay, M. J. (1991 [1976]) Norms and Ideology. In: Sociology of Science: A Sociological Pilgrimage. Open University Press, Milton Keynes, pp. 62-78.
  12. Woolgar, S. W. (1988) Science: The Very Idea. Ellis Horwood, Chichester.
  13. Woolgar, S. W. (1991) Beyond the Citation Debate: Towards a Sociology of Measurement Technologies and Their Use in Science Policy. Science and Public Policy 18: 319-26.
  14. Wouters, P. (1998) The Signs of Science. Scientometrics 41: 225-41.
  15. Zuckerman, H. A. & Merton, R. K. (1973) Institutionalized Patterns of Evaluation in Science. In: Storer, N. W. (Ed.), The Sociology of Science: Theoretical and Empirical Investigations. University of Chicago Press, Chicago, pp. 460-96.

Back to Top

Back to Sociology of Science