College and Research Libraries Characteristics of the Literature of Literary Scholarship Madeleine Stern The aim of the study was to determine how researchers use the literature that constitutes liter- ary scholarship. For this purpose three creative writers and three literary movements were selected for analysis. The 1976-80 volumes of the Arts and Humanities Citation Index pro- vided access to the articles written on the subjects chosen. The references cited by the authors of the articles, along with the form analysis of the articles themselves, constituted the substance on which the results of the study were based. The results support the contention that humanistic scholarship and the methods by which its characteristics are studied and analyzed are distinct and unique and require different meth- odology than that used for the analysis of information transfer in the sciences and social sci- ences. he purpose of this study is to try to determine how research- ers use the literature that con- stitutes literary scholarship. The assumption is that this type of schol- arship is characterized by unique attrib- utes and that its methodologies and re- search tools differ from those in other disciplines in the humanities and most certainly in the sciences and social sci- ences. The major thrust of literary scholarship is a critical concern for past and present creative writing. This critical concern may center around an author and his works, around the concept of critical theory itself, or around the formulation of literary theo- ries based on historical or cultural ap- proaches to criticism. The present study aims to focus on the two former aspects of critical concerns in literary scholarship, namely on research on creative writers and their work and on research on the theory and approaches to literary criticism. METHODOLOGY To examine how researchers in literary criticism use the literature on creative writers, and on the theory and approaches to literary criticism, three creative writers and three literary movements were se- lected for analysis. The authors chosen were John Milton, Henry James, and W. H. Auden; the literary movements were Symbolism, Existentialism, and Structur- alism. The 1976-80 volumes of the Arts and Humanities Citation Index (A&HCI) pro- vided access to articles written on the sub- jects. The references cited by the authors of the articles, along with the form analy- sis of the articles themselves, constituted the substance on which the results of the study were based. The Permuterm Subject Index was the starting point for the study. This index lists every significant English word in the titles of source articles indexed by A&HCI. "Beneath these 'primary terms' are listed alphabetically the 'co-terms' which ap- peared with it in titles. Opposite each 'co- Madeleine Stern is head, Collection Development, Lockwood Memorial Library, State University of New York at Buffalo. 199 ~\ 200 College & Research Libraries term' is the name of the author of the [source article] whose title includes the primary and the 'co-term'. " 1 While this method identified the creative writers and literary movements written about during the latter part of the 1970s, it did not neces- sarily identify all scholarly articles about these authors and literary movements in- dexed in the A&HCI. Nevertheless, it was felt that the appearance of the name of the author and of the movement in the title es- tablished in all instances a high degree of subject relevance, i.e., if an author's name appeared in the title it was more likely that the article was about the author than if his name did not appear in the title. Once the authors of the articles were thus identified, the "Source Index" sec- tion of A&HCI was consulted as a second step to the compilation of data for the anal- ysis. The "Source Index" is an author in- dex arranged alphabetically, and each en- try includes the name of all authors, full title of the article, the title of the journal, volume, issue, pages, and year of publica- tion. Beneath this bibliographic descrip- tion appears, alphabetically by author, the list of references cited by the author of the source article. This listing includes the cited author's name, the cited work or the title of the journal if the cited work is an article, the year of publication, and page number. The A&HCI differs from the SCI and the SSCI in that it goes beyond the practice of indexing only references cited in the bibliography or the footnotes of the source article. It also includes as cited items "works of art (novels, paintings, musical compositions, etc.) which are the subjects of the articles, even though they were not formally cited. " 2 A&HCI calls these "implicit" citations, and they are in- dicated by an arrow next to the cited title. From the "Source Index" entries, it was possible to determine whether the source article was a general article, a book review, · or a number of other types of publications. The list of the cited items were classified according to form into books, journals, and "other," which included manu- scripts, various communications, encyclo- pedias, dictionaries, theses, and illustra- tions. The cited items were divided according to content into primary sources July 1983 and secondary sources. In the study on creative writers, the primary sources were further divided into the works of the writer and other primary source materials. The publication dates of the cited items provided the data for the study on date distribution. Two problems arose when this method- ology was applied. Both of them can be at- tributed to the unique character of the lit- erature that deals with the works of creative writers and the method A&HCI employs to deal with them. By instituting the practice of "implicit" citations, A&HCI recognized the fact that creative works are the products of one person's ar- tistic and imaginative thought processes, and the scholarship that illuminates them and provides access to them cites and re- fers to them differently than when scien- tists and social scientists cite material rele- vant to the articles they write. A creative work may be quoted extensively within the text of the article, it may be mentioned in comparison with some other works, or referred to frequently within the text in connection with the thesis the author of the article wishes to expound. These refer- ences may appear as formal citations in footnotes or bibliographies in the same format. These may follow the same con- ventions used for citations in other aca- demic disciplines, or the author of the arti- cle may state in a footnote that all references to the cited works are to a par- ticular edition of the discussed writer's complete or selected works. Alternatively, the work referred to or quoted in the text of the article may not appear at all in for- mal footnotes or bibliographies. For these instances, A&HCI established the practice of indexing "implicit" citations. As a con- sequence of this indexing practice, the me- chanical aspect of indexing only what is formally cited is now expanded into a judgmental function, where the indexer has to scan the text and make a decision as to what to include as an "implicit" cita- tion. Thus, the subjective and interpretive element that characterizes literary scholar- ship, and differentiates it from scholar- ship in other disciplines, becomes incor- porated into humanistic citation indexing practices as well. In checking the source The Literature of Literary Scholarship 201 articles against the entries in the "Source Index,'' there were frequent questions as to why some references were indexed and some omitted. In spite of these ambigui- ties, both the formal and "implicit" cita- tions were counted and tabulated as they appeared in the "Source Index" beneath the bibliographic description of the source article. A related problem arose from the prac- tice of literary researchers citing collected works.A&HCI indexes these, sometimes under the name of the creative writer, and sometimes under the name of the editor. In some circumstances the individual works are also indexing terms, but not in all instances. In the study the citations of collected editions were always counted as primary sources under the creative writ- er's name, unless it was determined from the article that the reference was to the ed- itor's comments or opinions. In these cases the citation was counted as a second- ary source. The primary source count in the study on creative writers included the citations to the complete works as well as to the individual works, cited formally as well as implicitly. For each of the six subjects, three full or partial years of the A&HCI (between the years of 1976 and 1980) were utilized as a database. The bibliometric count dis- played in tables 1 and 2 indicates the num- ber of source items and the number of ref- erences counted, and the percentages of the totals for each subject. While the actual numbers of the counted items varied in size, the percentages showed general in- ternal consistency-a variance of not more than 10 percent-with the exception of the subgroup of secondary source references in the James, Auden, and Existentialism counts. For the age distribution count (see tables 3 and 4), the publication dates were di- vided into ten-year segments to 1920, one twenty-year span from 1920 to 1900, and one segment for all dates prior to 1900. A relatively large proportion of the cited items in the A&HCI lacked publication dates. This was partly due to omissions by the authors of the source articles, and partly due to omission of the cited date in the source article by the indexers of the A&HCI. Wherever the publication dates were ascertainable, they were added to the count and tabulated. Citations with unascertainable dates were not included in the age distribution analysis. THE WRITER AND HIS WORKS While ancillary topics, such as sources of a writer's work, his life, influence on other writers, comparisons with other au- thors' works, and relationships between his life and his writing have always been an integral part of literary research, the fundamental and predominant subject of investigation has remained the text itself. The study of the creative process, mani- fested by those who commit their thoughts and imagination to paper, is a historical, cumulative continuum. Writ- ings of the most contemporary authors be- come part of literary history once they are committed to the printed and published page. As students and scholars dissect and scrutinize these writings, a subfield of primary texts emerges and continues to expand. This subfield, a cross between primary and secondary source materials, comprises the large variety of publications beyond the first edition of an author's cre- ative work. This critical apparatus sur- rounding the creative process in literature adds another dimension to it, is the most distinctive feature of literary scholarship, and is shared to varying degrees by other humanistic disciplines. In selecting Milton, James, and Auden as subjects of these investigations, I in- tended to examine the influence of the time element on the corpus of research into the writings of authors separated from one another by centuries. Does the accumulation of works on Milton, as uti- lized by current researchers, differ in form and content from current research on the creative output of Auden? Does current Milton scholarship differ from current Auden scholarship? How does the accu- mulation of evaluative, revisionist, or other critical writings of the past influence the practitioners of present-day literary re- search? How important are primary sources? Is there any significant difference in the time span of these materials used by scholars? TABLE 1 CREATIVE WRITERS: SOURCE ARTICLE AND REFERENCE ANALYSIS N Sou rce Items References (Form ) to: Refere nces (Content) to: 0 Book Rev. Articles Books Journals Other Authors• Other ~imary Seco%dary N no. % % no. % % % no . % ]. Milton 1976 70 57 43 491 86.2 11 .8 2 491 18.5 26.3 55.2 (j 1979(Jan- Apr) 32 56.3 43.7 370 87 10 .3 2.7 370 27.8 18.4 53.8 0 1980(Jan- Aug) 72 55.6 44.4 739 83 13.5 3.4 739 14.9 26 59.1 = ~ Totals 174 56 .3% 43 .7% 1600 89.9% 12.2% 2.8% 1600 19% 24.3% 56.7% QQ ~ Au den ~ 1976 15 33.3 66 .7 99 76.8 23.2 2 99 51.5 12.1 36.4 ~ ~ 1978 21 33 .3 66.7 166 75 .3 21.6 2 166 38.5 10.1 51.2 (I) ~ 1979 (Jan-Aug) 11 36.4 63 .6 150 88 10 2 150 39.3 16 44.6 e: Totals 47 34% 66% 415 80.2% 17.8 2% 415 41.9% 12.8% 45 .3% n =- H. fames r-t ..... 1980 (Jan-My) 56 64 .2 35.8 413 81.6 16.2 2.1 413 24.9 18.4 56.5 0'" lot 1977 25 44 56 221 84.6 15.4 221 24.4 13.1 62.4 e: 1979 (Jan-Apr) 25 60 40 227 84.1 13.7 2.2 227 42.3 9.7 48 ;;· Totals 106 58.5% 41.5% 861 83% 15.3% 1.6% 861 29 .4% 14.7% 55.9% Cll • Auth ors refers to Milton, top; Auden, middle; and James, bottom . ._ TABLE2 = -'< LITERARY THEORY: SOURCE ARTICLE AND REFERENCE ANALYSIS ~ \C) (J) Source Items References (Form) to : References (Content) to : ~ Book Rev . Articl es Books Journals Other PriW,ary Seco%dary no. % % no . % % % no . S~mbolism 1 76 58 32.8 67 .2 818 73.6 13.8 12.6 818 30.3 69.7 1978 38 39.4 60 .6 595 73 .6 17.3 9.1 595 33 67 1979 (Jan-Apr) 42 33.3 66.6 570 78.8 15 6.1 570 28 72 Totals 138 35.1% 64.8% 1983 75.1% 15.2% 9.7% 1983 30.5% 69.5% Structuralism 1976 70 60 40 575 80.6 16.9 2.6 575 9.7 90 .3 1978 48 66 .6 33.4 698 77 22 1 698 7.9 92 .1 1979 (Jan-Apr) 18 61 39 27.0 73.3 25.1 1.5 270 7 93 Totals 136 62 .5 37.5 1543 77 21.3 1.7 1543 8.4 91.6 Existentialism 1976 36 47 .2 52.7 250 84 12.4 3.6 250 20.8 79.2 1978 26 34.6 65.4 275 84.4 15.6 275 29 .4 70.6 1979 (Jan-Apr) 16 37.6 62.5 93 73 14 6 43 43 57 Totals 78 41% 59% 618 83.3% 14.3% 2.4% 618 28% 72% The Literature of Literary Scholarship 203 TABLE 3 CREATIVE WRITERS: AGE DISTRIBUTION OF ALL REFERENCES % OF REFERENCES Milton James Au den Cum. Cum. Cum. Total Cum. % % % % % % % 1971- 23 23 21.5 21.5 20.4 20.4 21.6 1961-70 14.4 37.4 23.7 45.2 31.1 51.5 44.7 1951-60 6.4 43.8 10.8 56 13.6 65.1 55 1941-50 3.6 47.4 5.9 61.9 9.9 75 61.4 1931-40 4.2 51.6 3.2 65.1 21.2 96.2 71 1921-30 2.2 53.8 1.1 66.2 3.1 99.3 73.1 1901-20 1.8 55.6 13 79.2 0.5 99.8 78.2 to 1900 44 99.6 20.6 99.8 99.7 TABLE 4 LITERARY THEORY: AGE DISTRIBUTION OF ALL REFERENCES % OF REFERENCES Symbolism Cum . Existentialism Cum . Structuralism Cum. Total Cum . % % % % % % % 1971- 1961-70 1951-60 1941-50 1931-40 1921-30 1901-20 to 1900 21 21 25.8 25.8 44 44 30.3 58.5 70.4 75.2 79 .2 81.7 86.1 99.7 19.8 40.8 31 56.8 34 78 9.9 50.7 17.5 74.3 8.7 86.7 4.2 54.9 7.3 81.6 2.4 89.1 2.9 57.8 6.3 87.9 2.9 92 2.9 60.7 2.3 90.2 2.3 94.3 8.4 69.1 2.8 93 2 96.3 30.9 100 6.5 99.5 3.8 100.1 Source articles that dealt with the three authors were of two types: book reviews and articles. From the start of the investi- gation, the large number of book reviews was quite apparent, a clear indication in it- self of the book-oriented nature of literary scholarship. More than one-half of the source articles about Milton and James were book reviews, 56.3 percent and 58.5 percent, respectively, and 34 percent in the case of Auden. These figures don't in- dicate that each review article reviewed a different book. The same book would be reviewed in several articles and there may be a correlation between the importance of the book and its author and the number of times it was reviewed. Nevertheless, the preponderance of book reviews as a tool of literary scholarship seems to be one of the characteristics of the discipline; it also in- dicates the importance literary scholars at- tach to the monographic form for publish- ing research results. The markedly lower proportion of book reviews as source arti- cles on Auden, a mid-twentieth century author, would suggest that contempora- neity is a factor in the size of monographic literature being currently written about writers or poets. Auden' s work has not been around long enough to elicit mono- graphic treatment from scholars to the same degree as that of the likes of Milton or James. For the 327 source articles dealing with Milton, James, and Auden, the A&HCI listed a total of 2,876 references. (These do not necessarily correspond to the formal footnotes or bibliographic notes in the arti- cles.) Of these, 82.7 percent were refer- ences to books, 15.1 percent to articles, 2.2 percent to other types of materials such as dissertations, encyclopedias, unpub- lished communications, and a few manu- scripts. Jones, Chapman, and Woods in their study found that even though archi- val records are the historian's primary re- search sources, manuscript and other unpublished materials were cited surpris- ingly infrequently. In their study, 12.6 percent of all references were to unpub- lished material, and 87.4 percent per- tained to published material. 3 The insig- 204 College & Research Libraries nificant number of manuscripts used by the authors of the.source articles may be accounted for by the fact that the research into the topic of an article is done in a less in-depth manner than if the published study were a full-length book. I suspect that analysis of citations in source mono- graphs dealing with the same subject would yield a higher percentage of un- published primary source citatjons. The prominence of monographic litera- ture as it pertains to contemporary literary scholarship in the area of creative writing is unmistakable. While previous studies on overall patterns of literature use in the humanities clearly indicate that the bulk of references are to monographs, literature dealing with the works of creative writers shows the highest percentage of mono- graph use. One must examine these fig- ures with some caution, however, be- cause the definition of book or monograph in the analysis of the forms of material cited might vary from study to study. Broadus, in his survey of citation studies, cites Vaughan, who found in music that 69.5 percent of the references were to ''mono- graphs" and Simonton, who in his study of fine-arts literature shows that 71.4 per- cent of the citations were to "books. " 4 Heinzkill, in his study of citations in En- glish literary journals, found that 75 per- cent of all citations were to books. 5 The source articles dealing with Milton had 84.9 percent of their references to books, 12.2 percent to articles, and 2.8 per- cent to other forms. Those on James had 83 percent of their references to books, 15.3 percent to articles, and 1.6 percent to other forms. In the case of Auden, 80 per- cent of the references were to books, 17.8 percent to articles, and 1.9 percent tooth- ers. The time span that separates current scholarship from the subject studied seems to be a factor here. Knowledge and scholarship in literary studies is cumula- tive. Works of scholars of previous de- cades or centuries are not superseded or discarded. The same literary text may be the subject of investigation and of textual analysis over and over again. The appa- ratus criticus necessary for the preparation of definitive editions, manuscripts, let- July 1983 ters, and other communication are all characteristic of humanistic research, and of literary research more specifically. As the citation analysis suggests, the book is the form through which this type of schol- arly work is presented and because of its cumulative and encompassing nature, the further the subject of the scholarly investi- gation extends into the past, the heavier the investigator's reliance on the book form will be for study. While the major lit- erary form cited by scholars is the book for all three authors, there is a heavier reli- ance on books in the case of Milton than in the case of the twentieth-century poet Au den. Heavy reliance on primary source mate- rials is one of the most distinctive charac- teristics of literary research in general, and for the writings of individual authors in particular. In the study of recent articles about Milton, James, and Auden, 47.4 percent of all the references were to pri- mary materials; 43.3 percent of the refer- ences to Milton, 44.1 percent to James, and 54.7 percent to Auden were primary material. It would seem that the more con- temporary the author, the greater is the use made of primary sources by scholars to investigate, study, and evaluate the au- thor's work. The more contemporary the author, the fewer are the number of criti- cal works available to researchers for their studies. By virtue of the paucity of pre- vious research into their work, research on contemporary authors must depend more on the texts themselves. In the case of a literary figure of the nineteenth or sev- enteenth century, the work has already been studied, dissected, documented, and evaluated by previous generations of literary scholars. A closer analysis of the primary materials used for this study will further confirm this characteristic. Pri- mary material, as cited in scholarly articles on literature, can be divided into works of the authors who are the subject of the source article, and other primary works that are the creative products of other au- thors. In the source articles on Milton, this division indicates that 19 percent of the references to primary materials were to Milton's works and 24.3 percent to the worksofotherauthors; inJames' case29.4 The Literature of Literary Scholarship 205 percent were to his work and 14.7 percent to the works of other authors; in the arti- cles about Auden 41.9 percent of the refer- ences to primary materials were to the works of Auden and 12.8 percent to other authors. The 19 percent of references to Milton's works and the 41.9 percent to Auden' s might possibly indicate a paucity of output by Milton as opposed to rela- tively greater prolificity by Auden. I don't believe that this is a matter of quantity. The single work Paradise Lost may be cited and quoted line by line, chapter by chap- ter more frequently than dozens of Au den's poems . The difference lies in the manner in which researchers utilize mate- rials for their research. The more distant the creative writer's work, the larger is the body of secondary literature built around it, so that the researcher does not have to rely exclusively on the text itself in order to support his judgment or interpretation. In the case of references to primary sources other than the creative writer's works, the study shows a reverse of percentages. The lowest percentage, 12.8 percent, was for source articles dealing with Auden; the highest, 24.3 percent, dealt with Milton, with James-related articles showing 14.7 percent of the references to other primary sources . The same phenomenon may ac- count for this practice as for the above, where the reliance on secondary material is heavier for the distant author than for the contemporary one. In one sense this type of primary source material takes on the function of secondary material, where the works of other creative writers are scrutinized and compared with the work or works of the author who is the subject of the source article. Analysis of the publication dates pre- sented difficulties and required adjust- ments because of citation indexing prac- tices and occasional inconsistencies in the A&HCI. The citation index contains an un- reasonably large number of undated cita- tions, mostly to primary sources. In these instan<;es publication dates of the first edi- tion of the works cited were added and counted. Citations with unascertainable dates were not counted in the age distribu- tion analysis. A reference may also be a ci- tation to a complete, part of a complete, edited, or revised edition, in which case the publication date is later than that of the first edition. Consequently, the analysis of the age distribution reflects a combination of contemporaneous and later publication of primary source materials. For instance, in the source articles on Milton, 44 percent of all references were to publications prior to 1900 because a large percentage of the citations (43.3 percent) were to Milton and to other primary source materials first published prior to 1900. Yet some addi- tional citations were also made to later, re- vised editions, which are reflected in the distribution for later time periods. In the Milton studies 37.4 percent of all the cited material was twenty years old or less, while these figures are 45.2 percent and 51.5 percent for the James and Auden studies, respectively. The longer the time span between the creative writer's contri- bution and the work about him, the fur- ther back the literary scholar reaches for material . There is an indication here to the validity of statements about humanistic scholarship being cumulative and least susceptible to obsolescence. At the same time, it is interesting to note that current publications that consist mostly of second- ary sources, plus new editions of primary sources, seem to be important and at an equal level for all three. Twenty-three per- cent of the references on Milton were ten years old or less at the time the source arti- cle was written. These percentages for James and Auden were 21.5 and 20.4, re- spectively. The importance of contempo- rary material consisting largely of the writ- ings of the creative writer is reflected also in the median citation age of the refer- ences used in the source articles. Fifty per- cent of the references in the articles on Milton were published within fifty years of the date of the source article. Fifty per- cent of the references on James were pub- lished within approximately twenty-five years, and 50 percent of the references on Auden were published within twenty years. The least productive years on the pre- twentieth century authors seem to be the 1920s, with 2.2 percent and 1.1 percent of the references to Milton and James, re- spectively. Citations in the articles dealing 206 College & Research Libraries with Auden drop sharply after the high in the thirties (a decade of high productivity for Auden) with no citations dating from the pre-1900 period. The age distribution chart reflects the periods of significant productivity of three authors: 44 percent of all references in the Milton studies were to pre-1900 publications; James wrote in the late nineteenth century and the first two decades of the twentieth century, and the references for that period accounted for 33.6 percent of all references; Auden's contributions extended from the thirties, with the sixties being the peak decade with 31.1 percent of all references dating from that period, thus further confirming the importance of the creative work as pri- mary source material in this type of liter- ary research. LITERARY MOVEMENTS Another type of critical concern of liter- ary scholarship centers around the con- cept of critical theory espoused by literary movements and schools. The intent in this part of the study was to investigate by means of citation analysis (a) the survival rate in current literary scholarship of the doctrines that dominated the literary scene in several countries during a rela- tively short period of time, (b) to see if there are basic characteristics that separate movements from one another, and (c) whether citation analysis could pinpoint outside influences that shaped and/or dominated literary movements. Each of the three literary movements se- lected for the study had its roots in differ- ent disciplines, yet the span of time in which they dominated the literary scene was approximately the same-fifteen to twenty years. The proponents of the Sym- bolist movement of the latter part of the nineteenth century were mostly poets with a cult of beauty, and having close af- finities with the fine arts. Existentialism as a literary movement had its roots in philo- sophic doctrines of the mid-nineteenth and early-twentieth centuries, and its ex- pression flourished with a series of plays and novels during the forties and fifties. The principles of linguistics and cultural anthropology, and their application to lit- erary theory, form the basis of the Struc- July 1983 turalist movement in literature during the sixties and early seventies. "[Structural- ism] sees itself as a 'human science' capa- ble of identifying all the structures that make up a coherent universe." 6 The methodology used f