LEWIS CARROLL, writing under his real name, Charles Dodgson, suggested “that the proper definition of ‘Man’ is an ‘animal’ that writes letters”. A modern-day definition of scientists as “animals that write papers” might accurately reflect the central role of journals and publishers in influencing the practice of science. Science generally progresses by incremental steps. Our understanding of nature and the development of the diverse disciplines of science over the centuries have often involved decades of fruitless labour by dedicated scientists who have been denied the penetrating insights that accompany path-breaking advances. Nevertheless, their intense labour, faithfully recorded in research papers in scholarly journals, has often laid the foundation for the wonderful edifices of knowledge built by those who follow. Isaac Newton famously remarked: “If I have seen further than others, it is by standing on the shoulders of giants.” This sentiment underscores the importance of prior knowledge in facilitating even the most profound of scientific revolutions. Unsurprisingly, research journals are central to the practice of science. Since the first journal, Philosophical Transactions of the Royal Society London, appeared in 1665 the field of science publishing has grown dramatically. By the middle of the 20th century, it became clear that the explosive growth of research papers was beginning to make it difficult for scientists to keep abreast of even the most relevant publications in their chosen specialities. Nowhere was this problem more acutely felt than in chemistry, where an ever-increasing number of chemicals (molecules, in the parlance of the field) were being identified from natural sources and an even larger number were being synthesised in laboratories. The need to organise the literature of science and to draw connections between published papers was beginning to be felt by the 1950s.
Seminal paperIn 1955, a paper with the unassuming title “Citation Indexes for Science: A New Dimension in Documentation Through the Association of Ideas” appeared in the journal Science (Vol.122, pages 108-111, 1955), the flagship publication of the American Association for the Advancement of Science. This paper is now widely recognised as the starting point for the discipline of scientometrics, a field that attempts to quantitatively measure and assess scientific progress. Eugene Garfield, the author of the paper, died in Pennsylvania, United States, on February 26 at the age of 91.
In the years that followed the publication of his seminal paper, Garfield developed the Science Citation Index; introduced a journal-ranking parameter, the Journal Impact Factor (JIF); and created the Web of Science, a database widely used today to map trends in science and to extract quantitative metrics to assess countries, institutions and individual scientists. In later years, the Institute for Scientific Information (ISI) in Philadelphia that Garfield founded also introduced the Social Sciences Citation Index and the Arts and Humanities Citation Index.
Eugene Garfield, born in 1925 as Eugene Garfinkle, came of age during the years of the Great Depression and the Second World War. He graduated with a B.A. degree in chemistry from Columbia University in 1949. Chemistry catalysed his interest in cataloguing information as the literature of the discipline was growing at an exponential rate. A brief period as a research assistant and potential graduate student in the laboratory of the physical organic chemist Louis Hammett at Columbia persuaded Garfield that his future lay not in the experimental sciences but in organising the vast literature of science in a manner that permitted an understanding of the interconnectedness of research publications. A conversation on the sidelines of a session on “Chemical Literature” at an American Chemical Society meeting resulted in a position in a project at Johns Hopkins on medical indexing.
As Arnold Thackray and David Brock note in a brief biographical essay, the Army Medical Library (a precursor of the National Library of Medicine, which today runs PubMed Central, an indispensable literature search tool for biological and medical researchers) “was contending with the enormous expansion and increased pace of the medical literature, and hence was interested in developing new ways in which machines might be used to organise and search this dynamic body of new scientific knowledge. Because medical science uses chemistry as the lingua franca for many of its fields and specialties, someone with extensive chemical knowledge was needed to tackle issues of chemical nomenclature usage in creating a unified structure by which the medical literature could be organised and searched.”
The work done on this project was presented at a symposium entitled “The First Symposium on Methods in Scientific Documentation” in March 1953. A remarkable outcome of this presentation was a letter from William Adair, a retired vice president of the publishers of Shepard’s Citations, an invaluable index the American legal profession uses. Adair suggested that citation indexing might indeed be a fruitful way of organising the scientific and technical literature. Thackray and Brock suggest that this was Garfield’s eureka moment ( The Web of Knowledge: A Festschrift in Honor of Eugene Garfield , eds Blaise Cronin and Helen Barsky Atkins, Asis Monograph Series, 2000). Garfield moved back to Columbia for a master’s degree in library science; it was during this period that he put together the ideas eventually published in his path-breaking paper in Science in 1955. Creating a print version of a Science Citation Index in the mid 1950s would have posed almost insurmountable logistical and financial problems.
In May 1959, Garfield received an unexpected letter from the geneticist Joshua Lederberg, who had been awarded the Nobel Prize in 1958 at the remarkably young age of 33. Lederberg wrote: “Since you first published your scheme for a ‘citation index’ about four years ago I have been thinking very seriously about it and I must admit I am completely sold….” Garfield responded to this unanticipated expression of support from a most influential scientist: “I hope you won’t be embarrassed by a show of emotion, but your memo almost brought tears to my eyes. It then seemed that over six years of trying to sell the idea of citation indexes had not been completely in vain. You might be surprised how few people will take the time and trouble to scribble such a note….”
Lederberg recognised that the citation index would reveal “parent-offspring relationships of publications”, in the metaphor of genetics. The Citation Index is best described by Garfield: “Almost all papers, notes, reviews, corrections and correspondence published in scientific journals contain citations. These cite—generally by title, author and where and when published—documents that support, provide precedent for, illustrate or elaborate on what the author has to say. Citations are the formal explicit linkages between papers that have particular points in common. A citation index is built around these linkages. It lists publications that have been cited and identifies the sources of the citations. Anyone conducting a literature search can find from one to dozens of papers on a subject just by knowing one that has been cited. And every paper that is found provides a list of new citations with which to continue the search.”
Impact factors
The “impact factor” can be traced to Garfield’s original 1955 article: “Thus in the case of a highly significant article, the citation index has a quantitative value, for it may help the historian to measure the influence of the article that is its ‘impact factor’.” Years later, he introduced the JIF, a measure that has been used to rank journals in different subfields of science. Over the years, the JIF has become one of the most misused tools of research assessment, with individual performance often being judged on the basis of journals in which papers have been published, with little regard for the content and relevance of individual papers. In India, it is commonplace to talk of the average impact factor of a list of publications produced by an individual scientist or collectively by an institution, a meaningless metric that seems to please science administrators.
Garfield was prescient in warning over four decades ago: “When the ‘Science Citation Index’ was first proposed its major objective was to break the so-called subject index barrier. Out of this bibliographic experiment has evolved a historiographic and sociometric tool of major importance. Like most other scientific discoveries, this tool can be used wisely or abused. It is now up to the scientific community to prevent abuse of the SCI by devoting the necessary attention to its proper and judicious exploitation” ( Nature , Vol. 227, pages 669-671, 1970).
In the years before he began his work on the Science Citation Index, Garfield’s entrepreneurial streak led him to address an important but simpler problem that scientists confront: the task of reading the literature in their fields and identifying papers of interest even as the number of journals grew. Working in the time-honoured American tradition of a “garage operation”, he began to produce the periodical Current Contents , which was simply a photocopied version of the contents pages of all the journals in a field, with an alphabetically arranged index of authors and, most importantly, their addresses. In the mid 1970s, as a young researcher at the Indian Institute of Science, Bengaluru, I (along with many of my colleagues) used to eagerly await the arrival of the weekly issues of Current Contents . Journals would arrive months later by surface mail (“sea mail” in those lazier days). Postcards, decorated with colourful stamps (in the fond hope that the recipients were philatelists), would be dispatched by airmail with the expectation that reprints of the papers of interest would eventually arrive. Photocopying, email and the Internet lay far in the future. Current Contents carried a weekly essay by Garfield, and I soon found myself waiting eagerly for each new issue. He was an engaging writer who wrote of science, scientists and their publications with an insight that was unmatched.
It was in these columns, later collected and reprinted as The Essays of an Information Scientist (ISI, Philadelphia), that I first read of citation classics, JIFs and scientists whose work had a great impact on their fields, as judged by the number of citations that they accumulated. Nobel Prize winners and “Nobel Class” scientists featured in these pages, and Garfield induced many to write essays (“Citation Classics”), which provided a personalised and always readable account of their work. In a 2006 lecture to the European Conference on Publications in Medicine and Biomedicine held in Lund, Sweden, Garfield said: “Out of about one million or so scientists who have published to date, 10,000 can be considered to be ‘of Nobel Class’. Approximately 750 of them have won Nobel Prizes. While there are exceptions due to the vagaries of the subjective (non-random) selection process, Nobel laureates publish five times the average number of papers but their work is cited 30 to 50 times the average.”
Hidden gems
Garfield used citation analysis to identify path-breaking work in science, sometimes by scientists who were not widely recognised. In an April 1986 essay in Current Contents , he drew attention to the work of Sambhu Nath De of Nil Ratan Sircar Medical College in Calcutta (now Kolkata), who discovered that the cholera enterotoxin secreted by the pathogenic organism Vibrio cholerae was the causal agent leading to watery diarrhoea. De’s work, published between 1953 and 1959, resulted in a “paradigm shift” (a phrase famously introduced by Thomas Kuhn in his book The Structure of Scientific Revolutions ) in the thinking about cholera and its treatment. It was Garfield’s essay that prompted me to spend two years, 1988 to 1990, researching the literature of cholera and inducing the principal actors to write for a special issue of Current Science , which came out in July 1990. De died in 1986, “unhonoured and unsung” in India. At that time, when Garfield stirred me to dig deep into De’s work, I recalled Thomas Gray’s immortal lines in “Elegy Written in a Country Churchyard”:
Full many a gem of purest ray serene
The dark unfathomed caves of ocean bear
Full many a flower is born to blush unseen
And waste its fragrance in the desert air
In analysing the work of the Nobel Prize-winning astrophysicist Subrahmanyan Chandrasekhar (1910-95), Garfield discovered the highly cited work of Sivaramakrishna Chandrasekhar (1930-2004), of the Raman Research Institute in Bengaluru, on the discovery of discotic liquid crystals.
The scientific literature consists of a staggeringly large number of published papers; a report in the October 30, 2014, issue of the journal Nature estimates that as many as 58 million items were indexed in Garfield’s Web of Science. If even the first pages of each of these publications were stacked up, the pile would be as high as Mount Kilimanjaro, that is about 5,900 metres high. The hundred most cited papers would represent just one centimetre at the peak. The vast majority of published papers which have garnered only one citation each would constitute the foothills, over which a narrow and massive peak towered.
In a lecture in Sweden in 2006, Garfield presented an analysis of the scientific literature between 1900 and August 2005. Of the 38,163,319 papers indexed, only 19,938,769 (52.2 per cent) were cited even once; 41.8 per cent of the scientific publications recorded in the Web of Science were never cited. This is a sobering fact considering that this index is selective in the choice of journals it covers; 8,275,741 (21.7 per cent) of papers were cited less than five times. Only 21,385 (0.00056 per cent) papers were cited over 500 times. Garfield recognised that the scientific literature advances on a broad front of mediocre publications: “. . . the growth of science is dependent upon an accumulation of many ‘mediocre’ results that are produced by hard work”. He went on to add: “Long live the mediocrities. Without them how could there be geniuses?” ( Current Contents , November 4, 1970, Essays of an Information Scientist , ISI Press, Philadelphia, page 131, 1977).
Garfield is widely regarded as the founder of the field of scientometrics. However, in keeping with his passion for tracing the evolution of ideas, Garfield credits John Desmond Bernal, Vasily Nalimov and Derek J. de Solla Price as among the most influential originators of the field. He points out that the term “scientometrics” is the English translation of the title of Nalimov’s classic 1969 monograph Naukometriya(Journal of Informetrics , Vol. 3, pages 173-179, 2009). The tools of citation indexing Garfield developed became more powerful after the computer revolution and the advent of the Internet. Indeed, it is Garfield’s creation that is the Web of Science today, a commercial enterprise that has seen many owners. Its success has spawned competitors such as Elsevier’s Scopus and Google Scholar even as governments and science policymakers scramble for quantitative data on scientific research output.
Powerful toolThe growth of the Internet has resulted in new metrics to assess the impact of publications available online. Webometrics and “sitations” are emerging terms. Garfield pointed out that “web sitations” may be “harbingers of future citations”. He cautioned that “one must distinguish between readership or downloading and actual citation in new published papers” ( The Journal of the American Medical Association , Vol. 295, pages 90-93, 2006). Garfield’s original view that the Citation Index would be a powerful tool for historians and sociologists tracing the genesis of ideas in science is one that is rarely discussed. Garfield was, in many ways, the first entrepreneur of the information age. He brought to his ventures a degree of scholarship that was unique. The citation-ranking schemes he introduced were the conceptual forerunners of the Google Page Rank algorithm. Garfield’s impact on the world of science and his influence on the sociology of science are immeasurable.
P. Balaram is professor emeritus and a former Director of the Indian Institute of Science, Bengaluru.
He was the editor of the Indian scientific research journal Current Science from 1995 to 2013 .
