The standard toolkit of the modern indexer includes a computer, indexing software and a printer. Twenty years ago, it was a pen, a box of index cards and a typewriter. In the middle of the sixteenth century, before sitting down to his task, Conrad Gesner[1] assembled the following items: glue, scissors, string, paper, pen and ink. We know these materials were associated with the indexer's craft from the lengthy discussion on indexing techniques contained in Gesner's Pandectae, published in 1548. He begins with a brief introduction noting the importance of indexes:
It is now generally accepted that copious and strictly alphabetically arranged indexes must be compiled, especially for large, complex volumes, and that they are the greatest convenience to scholars, second only to the truly divine invention of printing books by movable type. (...) Truly, it seems to me that, life being so short, indexes to books should be considered as absolutely necessary by those who are engaged in a variety of studies . . . whether one will be reminded of something one has read before, or so that one might find something new for the first time.[2]
He also discusses an issue unsettled to this day (whether one index or several is better) and one now largely resolved (whether foliation, pagination, or a chapter-paragraph system should provide the locators). A large part of the explanation, however, deals with the mechanics of the job. The practitioner should "excerpt" key items from the text, write them on a piece of paper in any order and indicate which word will later be used to alphabetize them. Gesner then presents a detailed explanation of how to use string, glue, sheets of paper and paper strips to form a device to hold the index slips. Once the slips are cut up, this device will allow easy rearrangement of the slips as the process proceeds.[3]
This detailed exposition indicates that some people at least already considered indexing an important task worthy of a great deal of thought and effort. By this point in time, the printed index had a history of some eighty years, although its intellectual foundations reached back into antiquity. In this discussion, we will look briefly at those foundations, and then consider in detail the nature and development of indexing during the first 150 years of printing. We will accord special attention to items in the UW Libraries collection, which illustrate many of the trends and techniques of the craft during these early times.
*****
When did the first index appear? In order to answer this question, we must define the term. For the purposes of this paper, we can consider an index to be a summary of the contents of a book, arranged in alphabetical order and providing page numbers, folio numbers or some other location indicators.[4] Thus, three elements must come together to create this reference tool so common in modern times: summary, alphabetization and locators.
Early in the second millennium B.C.E., writers were already producing summaries of their own or others' work: the clay envelopes to cuneiform documents sometimes provide an abstract of the contents.[5] The alphabet was used to arrange information at least as early as the third century B.C.E.[6] For example, a Greek inscription from that era, found at Halasarna on the Aegean island of Kos, has the text of a resolution concerning the participants in the cult of Apollo and Heracles. When the list of participants is revised, it says, "Let them keep the names inscribed on a whitened tablet according to letter in order from alpha."[7] As to locators, a simple form of this access tool is provided any time a work is divided into chapters. There are ancient examples of this: Pliny and Aulus Gellius, for instance.[8] However, the general lack of such divisions in antiquity made access to the information in papyrus rolls quite difficult.[9] We do find an early such system in the work of Eusebius of Caesarea in the fourth century C.E. He improved access to the Gospels by dividing them into sections and numbering them.[10]
The fifth century C.E. produced the first known example of these three elements arrayed together: the Apothegmata, a list of the sayings of the Greek fathers, includes a subject index.[11] Then, in the eighth century, John of Damascus prefixed an index to his Sacra parallela, a collection of Bible verses and sayings of the fathers. The following note explains the system of alphabetization, no doubt a novelty to his readers:
In order that the reader may readily and easily find what he is looking for, the index of the subject headings, i.e. the titles, is composed alphabetically and every subject heading that is looked for will be found in the text under the letter with which it begins.[12]
However, these seem somewhat isolated efforts. The first sustained culture of index-making accompanied the rise of universities and "scholastic disputation" around the thirteenth century C.E.[13] For locators, these books typically relied on numbered divisions of the works in question, not on page or folio numbers.
Thus, while the incunabula period is the infancy of printing, it is not the infancy of the book. Indexing, like many other aspects of bookmaking, had a culture of practice from the manuscript period to draw from and build upon.
*****
Given this history, it is not surprising that within fifteen years of the first European printing with movable type, a printed index makes its appearance - this from the same men who sponsored Gutenberg's efforts. In 1467, or sometime before, Johann Fust and Peter Schoeffer published St. Augustine's De arte praedicandi. It includes an index for which the anonymous editor and author of the preface, apparently a clergyman, takes credit.[14] Like most early indexes, this indexer relied not on page or folio numbers. It was not until the last decade of the fifteenth century that indexes based on foliation started to appear.[15] In this volume, the printers marked each paragraph in the margin with a set of twenty-five symbols: the 23 Latin letters from A-Z, followed by two additional signs, the common abbreviations for et and con. Then they recommenced the system from AB onward (skipping AA), then from BC on (avoiding BA and BB), etc. It is to chapter divisions and to these paragraph symbols that the index refers. The book provides seven pages of index to 29 pages of text, or 24.1%, a generous length by today's standards. This is explained in part by the lengthy phrases and sentences that make up the entries. Following common practice, the index was alphabetized only through the first few letters.[16]
Nicholas of Lyra's Moralia super totam bibliam, published by Paulus de Butzbach in Mantua, appeared fourteen years after the Fust and Schoeffer index and, apparently, contains the oldest printed index in the UW collection.[17] Nicholas, who lived from about 1270 to 1349, presents bible commentaries in this book, ordering his comments according to the organization of the Bible itself. The text, exclusive of front matter, runs 511 pages, to which the indexer has supplied an index of 23 pages.[18] Thus, the length of the index represents about 4.5% of the length of the text. This varies little from modern American practice, where indexes typically represent 5-8% of the book length.[19]
The preface to this volume prominently mentions the index in its title[20] and informs us that Brother Ludovicus a Turri acted as both editor and indexer. At the back of the book, the following heading precedes the index: "Incipit tabula miro artificio contexta super prenominatus opus."[21] Again, the alphabetization is only partial. Generally, the indexer considers the first two to four letters. For instance, "Joha~nis" precedes "Job"; "Meritu~" comes before "Meretricu~."[22] The entries represent sentences of some five to fifteen words, summarizing a section of the book. None of the entries verified was a direct quote from the book. For instance, the heading "Amici Job significa~t illos qui iniuste iustis in crepare cona~tur. Job.iiii. 185"[23] summarizes a passage of several sentences on the topic of Job and his friends.
The index refers to abbreviated book titles and chapter numbers to orient the reader. For instance, the heading "Adam creatio (...) Hiere.xviii. 329" refers to the 18th chapter of Jeremiah. In the 1540s, Gesner still felt this was the best locator method, since it allows "all editions of a work in whatever form [to be] compatible."[24] Indeed, if this practice were reinstituted today, it would avoid the problem of last-minute changes to a few pages of a book rendering the entire index inaccurate and would also permit the indexing of long Web pages which lack pagination.
Despite Ludovicus' best efforts, however, it would appear that an early owner of this volume found the locator system wanting and set out to give the work page-number access. First, he or she numbered each page of the volume by hand from 1 to 517 (inadvertently omitting the numbers 228, 325 and 496). The owner then began the laborious process of adding page-number locators to each index entry, writing the following explanatory note above the printed index headnote: "Primus indicat numerus caput, paginum vero~ secundus"[25] Good intentions notwithstanding, the do-it-yourself indexer only made it part way through the A's before abandoning the task.
In addition, on the blank page opposite the first sheet of the index, this presumed early owner handwrote a brief index to the names of the books of the Bible, entitled "Tabula libroru~ hoc in opere contentorum."[26] Perhaps, after concluding it was too much effort to add page numbers to the entire index, the owner decided this table could serve to get to the beginning of each book more quickly while still using the existing locator system. This owner, then, is part of a long line of dissatisfied customers, stretching back to the manuscript period[27] and up to Thomas Jefferson[28] and beyond, who found the access to the books they have purchased so inadequate, correction was necessary.
As to the identity of these indexers, these workers generally toiled in anonymity, as they do today. However, in the 83 incunabula with indexes which Wellisch inspected, he did find a few that identified the indexer. He notes that "indexers whose names have come down to us either cite their monastic order or indicate that they are learned men."[29] Indeed, clerics penned the two indexes discussed so far. In addition, prefatory matter and index headnotes sometimes indicate that the author himself supplied the index. And very often, the printer may have done the indexing, along with performing myriad other tasks.[30] No doubt, men formed the vast majority of these indexers. However, like in the printing profession at large and the earlier scribal culture,[31] a few women were doubtless at the task as well. There is little indication that indexing formed a separate profession; rather, it was one of many tasks that people active in the world of scholarship and printing fulfilled.
*****
So, we see that indexes appeared early in the history of printing. But just how prevalent were they in the first years of printing? A precise answer to this question is clearly beyond knowing, given the small percentage of early printed books that survive.[32] However, investigations of various scholars provide some initial clues. Rabnett reports that of 82 incunabula at the University of Toronto, ten have indexes (12.2%)[33] Wellisch reports that he examined 970 incunabula, of which 83 (8.6%) contained indexes.[34] Clearly, a substantial body of books included indexes.
To probe this question further, let us consider the large corpus of early books represented in Online Computer Library Center (OCLC) catalog. This database represents the collections of hundreds of libraries around the world. It has records for some 100,000 European titles for the period we are considering. For any number of reasons, one cannot say that the figures returned from this database give an exact portrait of the number of titles held which have indexes, much less a precise image of indexing in the fifteenth and sixteenth centuries.[35] However, the general trends revealed seem to accord with other observed facts and to represent real phenomena of early publishing. Short of a labor-intensive ocular examination of thousands of early books, this seems the most effective way to garner clues as to the development of indexing across time and geography.[36]
The following figures, which compare records for a variety of prominent publishing centers throughout Europe, were obtained by searching OCLC Worldcat with the publication type set to books ("bks"). Various forms of the city names were given in the publication place field in order to capture as many of the books published there as possible. For instance, Paris was searched as "paris* or parrhis* or lutece or luteti*." Years were restricted as indicated, arranged into 24-year periods that avoid rounded year numbers, where many records for books of uncertain date are grouped. The phrase "index or indexes" was inserted into the note field of the query, to obtain estimates for the number of books with indexes.
OCLC records indicating presence of indexes
| 1451-1474 | 1476-1499 | 1501-1524 | 1526-1549 | 1551-1574 | 1576-1599 | total 1451-1599 | |
| Antwerp | * | * | 0.6%--154[37] | 5.5%--578 | 7.8%--1443 | 10.8%--1125 | 7.8%--3533 |
| Cologne | * | 6.3%--396 | 5.8%--240 | 14.9%--538 | 17.6%--517 | 14.6%--521 | 12.4%--2443 |
| Nuremberg | * | 3.8%--343 | 1.9%--426 | 2.7%--629 | 3.9%--228 | 5.4%--185 | 3.1%--1980 |
| Strasbourg | 4.1%--123 | 7.1%--603 | 4.4%--987 | 8.8%--707 | 12.8%--336 | 10.5%--258 | 7.3%--3214 |
| Basel | * | 3.4%--356 | 8.2%--910 | 14.1%--1669 | 18.5%--1701 | 15.3%--760 | 13.9%--5657 |
| Venice | 4.9%--143 | 4.3%--2722 | 5.1%--1947 | 6.6%--2085 | 11.6%--3563 | 10.5%--3554 | 8.2%--14,615 |
| Seville | * | * | * | 3.7%--109 | 6.7%--149 | 9.6%--115 | 5.7%--527 |
| Paris | * | 2.3%--435 | 9.5%--1513 | 11.3%--2724 | 8.2%--2790 | 8.2%--1931 | 8.9%--9844 |
| London | * | 1.1%--187 | 6.4%--645 | 10.0%--2140 | 9.4%--3125 | 6.9%--4970 | 8.1%--11,617 |
| European totals[38] | 2.3%-- 1154 | 3.7%-- 9813 |
4.4%--
13,895 |
7.7%--
18,809 |
9.6%-- 24,542 | 9.3%-- 26,828 | 7.6%--100,283 |
*fewer than a hundred examples
While these figures don't give a us a precise image of early book indexing, they do permit us certain reflections. Indexes were less of a standard accompaniment to books in that period than they are today. For the year 1995, about 24.0% of books in six modern languages seem to have indexes,[39] compared to the figure of 7.6% listed above. And the modern corpus would include a large number of fiction titles, where an index has much less utility and is generally absent. Nevertheless, indexes appear to have been far from rare in the earlier period -- at least in the type of substantial, serious volumes likely to survive vicissitudes of taste and hardships of the elements through five centuries.
In addition, the table indicates that indexes gained in prevalence through the first century after Gutenberg, as printing itself was expanding briskly.[40] Indeed, publishers seem to have been aware of the value of a good index to book sales. Peter Schoeffer's prospectus cites his "more complete and better arranged" indexes as an important selling point.[41] In the 1520s, Wellisch notes, herbals began to advertise the inclusion of an index more prominently. A 1527 title page proudly proclaims "Also a new index in which one can see the whole contents of this book (as in a mirror)."[42]
These advances coincided with improvements in a variety of indexing techniques. By 1550, Europeans were offered elaborate indexes that in many ways surpass standard indexing practices of today. Some of the more elaborate examples included indexes rendering names in Latin, Greek and Hebrew script. Some indexes, including one from 1545, went so far as to indicate to readers both the page and the line on which a topic of interest could be found.[43] A 1544 index from the UW collection presents a lesser example of these efforts at precise access.[44] The index refers to folio numbers followed by the designations A-K. A-E are on the recto of the folio; F-K on the verso. Thus, the reader can identify which fifth of the page contains the item of interest. At times, however, the attention to detail can lead to confusion. A 1526 volume from the UW collection presents no fewer than nine separate indexes.[45]
We also notice that indexes underwent a decline in the latter half of the sixteenth century. This corresponds with the general decline of the quality of printing during that time of warfare and economic problems. Books were made more on the cheap. The quality of paper declined. Fewer illustrations were used; and those that did appear were often from old and worn woodcuts.[46] Indexing, a time-consuming and therefore expensive endeavor; could not escape the hard times affecting the business in general.
Information on the geography of indexing also emerges from these data. Gesner makes the comment that the Germans "have surpassed the French, Italians and Spaniards, because the latter's books have for the most part no indexes at all, and there are only a very few that have all too short ones."[47] The OCLC data seem to bear out these contentions, at least somewhat. Certain German-speaking publishing centers - Cologne, Strasbourg and Basel (Gesner had lived in the latter two)[48] - stand out as cities devoted to indexing. However, the German city of Nuremberg lags far behind. And, the Italians of Venice seem to rival some of the Germans in this matter.
Wellisch notes that the English trailed the Germans and French in the quality of indexes to their herbals.[49] This specific comment may apply to London indexing as a whole, which lagged behind the German and French centers of publishing according to the table above.
Finally, the OCLC data tells us that indexing in the vernacular was rarer by far than indexing in Latin. For the entire period 1451-1599, 10.2% of the records for Latin-language books indicate an index. The figures for the vernaculars range from 5.5% for Italian to 3.0% for German. This also seems to indicate that Gesner's high assessment of German indexing may not apply to regions far from the large centers with which he was familiar.
*****
As examples of these poorer relations to Latin indexing, let us examine some early French- and English-language indexes from the UW Libraries. In 1565, André Wechel of Paris published the text of the New Testament, excluding the Gospels and Acts, in a bilingual French-Latin volume. It bears the title Les épistres Sainct Paul, les épistres catholiques de Sainct Iaquis, Sainct Pierre, & Sainct Jean, l'Apocalypse ou révélation de Sainct Iean.[50] The French-language index is entitled "Table du nouveau testament" and does, indeed, provide references to the entirety of the New Testament, including the Gospels and the Acts of the Apostles. Apparently, this volume belongs to a two-volume set, which has been separated.
The book contains 484 pages of text, to which an index of 21 pages is appended.[51] Thus, the length is 4.3% that of the text. The index is in French and presents more succinct entries than Ludovicus' index to Nicholas, generally one to five words forming a verb phrase or noun phrase rather than a sentence. Despite the presence of printed folio numbers, the book relies on the book and chapter structure of the Bible to give users access to the text, with an added system of printed letters in the margin to orient the reader more specifically. For instance, the heading "Avarice, racine de to' maux, I.Tim.6:b" leads the reader to the sixth chapter of Paul's first letter to Timothy; then the reader finds the letter B in the margin to get closer to the actual section in question. In this case, one finds the verse stating "Car la racine de tous les maux c'est avarice." Despite the gains of foliation and pagination, the chapter-division technique -- Gesner's preferred locator system -- still had currency some 70 years after the first indexes based on foliation.
Alphabetization had progressed through the sixteenth century, as exemplified by a 1529 index published in Cologne which is an early example of strict alphabetical order.[52] Gesner applauded this trend, commending Nicolaus Erythraeus's index to Virgil for being "strictly alphabetical."[53] The Wechel Bible bears the mark of that trend. Generally, the entries are fully alphabetized through the first word, although later words are ignored. Each alphabetical block with different initial letters is separated by a space. Then, each block defined by a different second letter is marked by a paragraph sign. For instance, after the ca's, the first entry beginning with ce (Cène du Seigneur) is indicated by a paragraph mark. Interestingly, however, the alphabetization often is based on a keyword that is not the first word of the entry. For instance, "Iuifs Zélateurs de la Loy" is under the letter Z: "Toute chair comme l'herbe" is placed under the letter C.[54] This book contains no other front or back matter to indicate who created the index.
As with printing in general, indexing arrived late in Great Britain. We have already noted Wellisch's comment on the inferior quality of early English indexing of herbals.[55] And of the 83 indexed incunabula he identified, none were from England.[56] Therefore, it's not surprising that the UW Libraries seem to have no English-language indexes before 1600, outside of microforms and photocopies.[57] We can, however, examine a photocopy of Edmund Bollifant's 1586 edition of Pierre de la Primaudaye's The French academie, translated from the French.[58] The book concludes with "A Table of the Principall Matters Contained in this Academie." The index runs to ten pages for a book that has 812 pages of text, excluding the front matter. Thus, the index is a mere 1.2% of the text length. The front matter provides no clue as to who penned the index.
The index has headings and subheadings, consisting of phrases or sentences that range from about five to fifteen words. The following complete heading gives the flavor of the indexer's style:
Youth: how the Romans taught their youth to forsake the follies of their first age, 567. examples of vertuous young men, 568. how the Persian youth was instructed, 263. two things to be respected in the institution of youth, 556. the common disease of youth, 559. sixe precepts requisite in the instruction of youth, 558
All but the final subheading above draw on summaries printed in the margins of the text, mostly quoting directly from those summaries. Following the trend, the index relies upon page-number locators. Alphabetization of the main headings is fairly strict, with the occasional lapse, such as "Philosophy" before "Philosopher." Subheadings are roughly in page-number order, but with a large degree of randomness in their arrangement.
Whether the inadequacy of this brief, yet wordy index reflects the French original or English invention is unclear. However, the UW Libraries do possess a later edition in the original French with a much more substantial index. The French text, published in Basel, provides an index of 74 pages to 772 pages of text, representing 9.6%.[59] Gesner would have been proud to see his native Switzerland continuing the strong indexing tradition he devoted his energy to improving.
*****
So, we see that indexing emerged from 150 years of innovation and continuing practice as a strong element of the reference repertoire. Many innovative techniques were tried. Some, like page-number locators caught on as inherently useful. Others, like page and line indicators probably proved too expensive and time-consuming to predominate. Indexing necessarily responded to outside forces. Economic realities first pushed printers to include indexes to distinguish their works and later caused a slight retreat as indexing weathered the same economic storm that battered the rest of publishing. But indexing also worked its own influence on the world around it, on the minds of readers and the content of literature. As Eisenstein argues, the rigor of indexing forced editors to confront inconsistencies in medieval texts and favored rearrangement of large compilations along criteria of "unity, internal consistency, and harmony." It served, for instance, to bring access and order to the body of public law.[60] If indexing's place had not been assured by the foothold it established in the manuscript era, its usefulness was certainly widely recognized by the end of the sixteenth century. And its role would grow only more prominent in the centuries that followed.
Copyright ©1998-2000, Frederick L. Brown. All rights reserved.
Emili, Paolo. Pauli Aemylii veronensis ... de rebus gestis francorvm libri decem (...). Parisiis: Ex officina Michaelis Vascosani, 1544. Provides an example of precise locator system.
Epistres Sainct Paul, les épistres catholiques de Sainct Iaquis, Sainct Pierre, & Sainct Iean, l'Apocalypse ou révélation de Sainct Iean, Les. Paris: André Wechel, 1565. This book seems to represent the oldest French index in the UW collection.
La Primaudaye, Pierre de (1586). The French academie: wherein is discoursed the institution of maners, and whatsoever else concerneth the good and happy life of all estates and callings (...). London: Imprinted by Edmund Bollifant for G. Bishop and Ralph Newbery, 1586. This photocopied book contains what seems to be the oldest English index in the UW collection, outside of microforms.
La Primaudaye, Pierre de (1587). Académie françoise, en laquelle il est traité de l'institution des moeurs, & de ce qui concerne le bien & heureusement viure en tous estats & conditions (...) Basle: Philemon de Hus, 1587. This is an edition of the French original on which the preceding translation is based.
Maffei, Raffaele. Commentariorum Vrbanoru Raphaelis Volaterrani, octo & triginta libri (...). [Parrhisiis]: Venundatur Ioani Paruo, Iodoco Badio, Claudio Cheuallio, & Conrado Resch [1526]. Presents a example of early multiple indexes.
Nicholas of Lyra. [Moralia super totam Bibliam]. [Mantua]: [Paulus de Butzbach], 1481. This book seems to represent the oldest index in the UW collection.
Anderson, M. D. "The Length of book indexes." The Indexer 5:1 (Spring 1966), 3-4. Provides estimates for index lengths in a variety of subject areas. Methodology is not clearly explained.
Bühler, Curt F. The Fifteenth century book: the scribes, the printers, the decorators. Philadelphia: University of Pennsylvania, 1960. Includes discussion of the sociological background of early printers and editors.
Daly, Lloyd W. Contributions to a history of alphabetization in antiquity and the Middle Ages. Bruxelles: Latomus, 1967. Provides detailed discussions of the use of the alphabet, including early alphabetical lists and Bible concordances.
Eisenstein, Elizabeth L. The Printing revolution in early modern Europe. Cambridge: Cambridge University Press, 1983. This exploratory work sketches topics for further research around the influence of print culture on the Renaissance, the Reformation and the rise of science. It provides useful discussion on the nature of the printing enterprise during the early print period.
Febvre, Lucien, and Henri-Jean Martin. L'Apparition du livre. Paris: Editions Albin Michel, 1958. Covers roughly the first 150 years of printing with detailed statistics and assessments of broad characteristics of early printing, as well as discussion of individual houses.
Helfer, Robert S. "Alphabetization." In Wayne A. Wiegand and Donald G. Davis, Jr., eds., Encyclopedia of Library History. New York: Garland Publishing, 1994, pp. 21-22. Useful synopsis of key events in the history of alphabetization.
Johnson, A. F. Selected essays on books and printing. Amsterdam: Van Gendt & Co., 1970. Includes a chapter giving a country-by-country overview of printing in the sixteenth century. Special attention is given to specific print houses, but larger trends are discussed. Also includes a chapter on the Italian book in the sixteenth century.
Katz, Bill. Dahl's history of the book. Metuchen, N.J.: Scarecrow Press, 1995. Provides an overview of book history, with many useful references and brief comments on points covered in this paper.
McLaughlin, Jack. "The Organized president." American Heritage (July/August 1991), 86-88. Discusses indexing methods used by Jefferson in handwritten indexes he added to books in his collection.
Online Computer Library Center. "OCLC Worldcat" [database]. Dublin, Ohio: OCLC, 1998. This union catalog of over 30 million records from hundreds of libraries around the world provides flexible searching capabilities. However, cataloging quality is problematic.
Pilet, P. E. "Gesner, Conrad." In Charles Coulston Gillispie, ed. Dictionary of Scientific Biography. New York: Charles Scribner's Sons, 1972, pp. 378-79. This brief biography of the man described as the Father of Bibliography concentrates on his work in natural science.
Rabnett, Mark (1982). "The First printed indexes: a study of index techniques in some incunabula." Cataloging & Classification Quarterly 2:3/4 (1982) 87-102. Examines the "structure, style and technique" displayed in incunabula indexes, with detailed description of several examples.
Wellisch, Hans H. (1978). "Early multilingual and multiscript indexes in herbals." The Indexer 11:2 (October 1978), 81-102. Examines a number of herbal indexes and traces the evolution of indexing in this field. The article includes numerous reproductions from the indexes.
----- (1981). "How to make an index -- 16th century style: Conrad Gessner on Indexes and Catalogs." International Classification 8:1 (1981), 10-15. Includes a lengthy translation of Gesner's fascinating comments on indexing, as well as discussing his ideas on library catalogs and the indexes to his works.
----- (1986). "The Oldest printed indexes." The Indexer 15:2 (October 1986), 73-82. Considers the claims of two early indexes to being the earliest printed index, describing the indexes and the printers who produced them.
----- (1991). Indexing from A to Z. Bronx, N.Y.: H. W. Wilson Company. Discusses various aspects of indexing, concentrating on information useful for the practitioner. Includes an unscientific estimate of standard index lengths.
----- (1994a). "Abstracting." In Wayne A. Wiegand and Donald G. Davis, Jr., eds., Encyclopedia of Library History. New York: Garland Publishing, 1994, pp. 1-2. Brief summary of key events in abstracting from the origins to the present.
----- (1994b). "Incunabula indexes." The indexer 19:1 (April 1994), 3-12. Examines a broad corpus of incunabula indexes, with detailed discussion of techniques, quality and relation to the printing business.
----- (1994c). "Indexing." In Wayne A. Wiegand and Donald G. Davis, Jr., eds., Encyclopedia of Library History. New York: Garland Publishing, 1994, pp. 268-69. Useful overview of important stages of indexing's development.
Witty, Francis J. (1965). "Early indexing techniques: a study of several book indexes of the fourteenth, fifteenth, and early sixteenth centuries." The Library Quarterly 25:3 (July 1965), 141-48. Discussion of indexes found in late medieval manuscripts and early printed books, including detailed exposition of indexing techniques.
----- (1973). "The Beginnings of indexing and abstracting: some notes toward
a history of indexing and abstracting in antiquity and the middle ages."
The Indexer, 8:4 (October 1973), 193-98. Provides a brief overview
of indexing history, including references to early examples of alphabetization
and document summary.
[1] I will follow the standard English-language
practice of spelling his name thus, although Wellisch argues that "Gessner"
is correct. "How to make an index -- 16th century style: Conrad Gessner on
Indexes and Catalogs," International Classification 8:1 (1981), note
1.
[2] The translations of Gesner in this paper of those of Wellisch (1981, p. 11).
[3] Gesner suggests the indexer prepare a sheet of paper which will hold two columns of slips. Eight strings will be run from top to the bottom, two on either side of each column, these strings being woven through holes at four points along the sheet -- at the top, the bottom and two equidistant points along the way. Next, two long strips of papers, each with an upturned edge turned toward the columns, are glued along the left and right edges of the paper to keep the slips from falling out. One strip of paper with upturned edges on both sides is glued between the two columns, to keep slips from sliding from one column to the other. Once a quantity of these sheets has been made, they can be bound together in a volume. Not only is this a useful method for compiling indexes, Gesner notes, but also for organizing notes as one sets about writing a book (Wellisch 1981, pp. 11-12).
[4] This is essentially the Oxford English Dictionary definition 5b: "An alphabetical list, placed (usually) at the end of a book, of the names, subjects, etc. occurring in it, with indication of the places in which they occur."
[5] Francis J. Witty, "The Beginnings of indexing and abstracting: some notes toward a history of indexing and abstracting in antiquity and the middle ages," The Indexer, 8:4 (October 1973), p. 193; Hans H. Wellisch, "Abstracting," in Wayne A. Wiegand and Donald G. Davis, Jr., eds., Encyclopedia of Library History (New York: Garland Publishing, 1994a), pp. 1-2.
[6] Robert S. Helfer, "Alphabetization," in Wayne A. Wiegand and Donald G. Davis, Jr., eds., Encyclopedia of Library History (New York: Garland Publishing, 1994), pp.21-22.
[7] Lloyd W. Daly, Contributions to a history of alphabetization in antiquity and the Middle Ages. (Bruxelles: Latomus, 1967), pp. 18-19.
[9] Bill Katz, Dahl's history of the book (Metuchen, N.J.: Scarecrow Press, 1995), p. 49.
[11] Witty (1973), p. 196; Hans H. Wellisch, "Indexing," in Wayne A. Wiegand and Donald G. Davis, Jr., eds., Encyclopedia of Library History (New York: Garland Publishing, 1994c), pp. 268-69.
[13] Witty (1973), p. 197; Witty, Francis J., "Early indexing techniques: a study of several book indexes of the fourteenth, fifteenth, and early sixteenth centuries," The Library Quarterly 25:3 (July 1965), p. 141.
[14] Hans H. Wellisch, "The Oldest printed indexes," The Indexer 15:2 (October 1986), pp. 73-82.
[16] Wellisch (1986), p. 81; Witty (1965), pp. 144, 146.
[17] Leaving aside books reproduced in the microform collection.
[18] With two columns to the page and approximately 57 lines to the column.
[19] Wellisch notes that the typical modern index for history, biography and high school and college textbooks is about 5-8% of the text. Indexing from A to Z (Bronx, N.Y.: H. W. Wilson Company, 1991), p. 211. A brief study by Anderson of twentieth-century books comes up with the same figure of 5-8% for history books, but puts most biographies in the 1-4% range. "The Length of book indexes," The Indexer 5:1 (Spring 1966), p. 3. Neither writer clearly identifies his methodology in obtaining these numbers. However, these figures accord generally with my experience working as an indexer.
[20] "Epistola pro operis emendatione & pro ipsius Tabula miro artificio ordinata." ("Note on the editing of this work and on its index, ordered with wondrous craft").
[21] "Here begins the index constructed with wondrous craft based on the preceding work."
[22] The tilde following a letter represents the symbol placed above that letter in the original to denote an abbreviated form of a word.
[23] The number in italics refers to a handwritten number added by an early owner. See discussion below.
[25] The first number indicates the chapter, the second, the true page.
[26] Index of books contained in this work.
[27] In several manuscript books, Witty notes hurriedly written indexes in a different hand than the rest of the book. He proposes that these may have been written by the owners (1973, p. 197).
[28] Jefferson compiled indexes to at least three books in his possession. Jack McLaughlin, "The Organized president," American Heritage (July/August 1991), pp. 86-88.
[29] Hans H. Wellisch, "Incunabula indexes," The indexer 19:1 (April 1994b), p. 9.
[30] Elizabeth L. Eisenstein, The Printing revolution in early modern Europe (Cambridge: Cambridge University Press, 1983), p. 61.
[31] Curt F. Bühler, The Fifteenth century book: the scribes, the printers, the decorators (Philadelphia: University of Pennsylvania, 1960), pp. 22, 50.
[32] Wellisch cites estimates of 27,000 to 36,000 for extant incunabula (1994b, p. 3), whereas Katz estimates that 20 million may have been published (1995, p.166). Although these are both estimates, that would mean less than 0.2% of the production of the period survives.
[33] Mark Rabnett, "The First printed indexes: a study of index techniques in some incunabula," Cataloging & Classification Quarterly 2:3/4 (1982), p. 89.
[34] The items were primarily from the Folger Shakespeare Library, the Dibner Library of the Smithsonian Institution, and the National Library of Medicine. A few items from the Georgetown University Library and the Library of Congress were also included (Wellisch 1994b, p. 3).
[35] A key problem is that records are submitted by member libraries, without effective culling of duplicates.
[36] Other large catalogs, such as the Library of Congress, the University of California's Melvyl, the Center for Research Libraries and Research Libraries Information Network, don't easily provide the flexible searching needed for these types of investigations.
[37] 0.6%--154 indicates that 154 books were found in this category, of which 0.6% had "index or indexes" in the note field.
[38] These totals are for books in Latin, French, Italian, Spanish, German, Dutch and English. Due to the limited search abilities of the database, figures for English are only from books identified as being published in London (which represent about 2/3 of all English-language titles in the database).
[39] This figure is for all the books in the database in French, Italian, German, Spanish and Dutch, but only the New York- and London-published books in English. New York and London represent only about a fifth of the English-language titles in the database.
[40] Lucien Febvre and Henri-Jean Martin, L'Apparition du livre (Paris: Editions Albin Michel, 1958), pp. 281, 285-86.
[41] Eisenstein (1983), p. 65.
[42] Hans H. Wellisch, "Early multilingual and multiscript indexes in herbals," The Indexer 11:2 (October 1978), p. 87.
[44] Paolo Emili, Pauli Aemylii veronensis (...) de rebus gestis francorvm libri decem (...) (Parisiis: Ex officina Michaelis Vascosani, 1544).
[45] Raffaele Maffei, Commentariorum Vrbanoru Raphaelis Volaterrani, octo & triginta libri (...). ([Parrhisiis]: Venundatur Ioani Paruo, Iodoco Badio, Claudio Cheuallio, & Conrado Resch [1526]).
[46] Katz (1995), p. 175. A. F. Johnson, Selected essays on books and printing (Amsterdam: Van Gendt & Co., 1970), p. 56. Febvre and Martin (1958), p. 145.
[48] P. E. Pilet, "Gesner, Conrad," in Charles Coulston Gillispie, ed., Dictionary of Scientific Biography (New York: Charles Scribner's Sons, 1972), pp. 378-79.
[50] This is apparently the oldest French-language index in the UW Libraries, outside of microform copies.
[51] Two columns per page, 34 lines per column.
[52] Dioscorides. In Dioscoridem corollarium libri quinque. (...) Coloniae: apud Joan. Soterem, 1530. (Wellisch, 1978, p. 88).
[54] Witty notes a similar practice in the index of the manuscript book, Commentarius in primum Sententiarum of Aegidius Romanus (1965, p. 142).
[56] Wellisch (1994b), pp.11-12.
[57] The earliest I identified was in Sir Walter Raleigh, Historie of the World (London: Walter Burre, 1614).
[58] The translator is identified as "T.B.C." in the preface. The UW catalog record indicates this may be "Thomas Bowes, Clerk."
[59] Pierre de La Primaudaye, Académie françoise (...) (Basle: Philemon de Hus, 1587).
[60] Eisenstein (1983), pp. 70-72.
Copyright ©1998-2000, Frederick L. Brown. All rights reserved.
updated September 9, 2000
return to home page