Incunabula indexes

Incunabula indexes
Hans H. Wellisch
Examination of nearly a thousand books printed before 1500 has revealed indexes, some printed,
some manuscript, in 83 of them, enabling some conclusions to be made on late medieval indexing
policy and practice. The likelihood of indexing in particular subject areas; whether the index {or
registrum or tabula) is explained; and the structure of the indexes, are among topics covered. The
vagaries of primitive alphabetization are outlined, and an attempt is made to assess quality. Some
of the indexers of these indexes are known by name. There is much worthy of study in medieval
indexes, which are not merely of antiquarian interest. A list of the examined items is given.
The limitation of the incunabula period to the first
fifty years or so of printing with an arbitrary cutoff
date at 1500 is in many respects unfortunate, as has
been stated by a noted historian of printing: This date
cuts right across the most fertile period of the new art,
halving the lives of some of its greatest practitioners'.
(Steinberg 1955, p 19). This is also true regarding the
indexes of incunabula. As this survey will show, these
indexes varied widely in quality and sophistication, but
a trend towards steady improvement is clearly dis
cernible toward the end of the period, and indexes
achieved higher quality just after the turn of the 16th
century. In the 1550s, editorial and typographical
features, including the compilation of multilingual
indexes, reached a level seldom seen in our modern
indexes. Still, a study of any aspect of incunabula
must necessarily adhere to accepted scholarly norms,
and this survey is therefore limited to the indexes of
books printed until the end of the year 1500.
Manuscript indexes are also covered, inasmuch as they
are relevant to those printed indexes that were
modelled on them.
The sample and its characteristics
The provision of finding aids in late medieval manu
scripts and in the earliest printed books has been the
subject of much uninformed speculation, such as that
manuscripts could not be indexed because they were
not paginated and no two of them were exactly he
same, or that printed indexes could only be compiled
after pagination had been invented. Only a few serious
inquiries have been published, but even those based
their findings on a small number of items that
happened to be available; (Pollard 1908; Witty 1965,
This article first appeared in Mercurius in trivia, a feslschrift
for the Italian librarian and bibliographer Alfredo Serrai,
edited by Maria Cochetti; Bulzoni Editore, 1992. It is repro
duced here by kind permission of the author. The reference
system adopted by the author in the original version is
retained here.
The numbers in square brackets refer to the list of indexed
incunabula in Appendix I.
The Indexer Vol. 19 No. 1 April 1994
1973; Rabnett 1982) or on a special subject (Wellisch
1978).
The present study is based on the personal inspec
tion of four fairly large and variegated collections of
incunabula, namely 330 in the Folger Shakespeare
Library, 17 in the Georgetown University Library, 340
in the Dibner Library of the Smithsonian Institution
Libraries (all in Washington, D.C.) and 258 in the
National Library of Medicine in Bethesda, Maryland:
a total of 953 books. In addition, a dozen items were
inspected in the Library of Congress, and some data in
the publications listed above were also used. All books
were checked for the presence of name and subject
indexes in addition to and separate from tables of con
tents (both of which are in most cases labelled
"Tabula" which makes their description in bibliogra
phies ambiguous, the more so since some genuine
indexes are sometimes printed right after the table of
contents in the beginning of books). Data were veri
fied or completed (in cases when parts of. an index
were missing) from those in the standard incunabula
bibliographies: Hain-Copinger's Repertorium and the
Gesamtkatalog
der
Wiegendrucke
and
Polain's
Catalogue des livres imprimis au XVe siecle des bibliotheques de Belgique.
The total of 970 books inspected represents roughly
2.8-3.6 percent of the total quantity of incunabula
which has been variously estimated as 36,000 (a figure
cited in the Encyclopedia Britannica) or as low as
27,000 (Dachs & Schmidt 1974). 83 books, or 8.6 per
cent of all books examined had either name or subject
indexes or both. The difference between the 83 items
seen and the 63 listed in the Appendix is caused by
reprinted indexes seen.
Almost all indexes were printed, but four books
[2, 28, 40, 50] had handwritten indexes, evidently com
piled by their owners who were frustrated by
the absence of an index: in one instance, a reader
compiled his own index in addition to a fairly large
printed one [40] which was apparently not sufficient
for his exacting needs.
INCUNABULA INDEXES
The subjects of indexed books
The earliest printed index, compiled for St.
Augustine's De arte praedicandi [8] was probably mod
elled on earlier manuscript indexes of the work; the
anonymous editor and indexer says in his preface that
he had collated the text of several manuscripts which
he had found in different places, though he may have
improved on his models because he also claims that it
had taken him a long time to compile the index. The
fact that this quite elaborate and sophisticated index
(which had double entries, cross-references and even
diagrammatic displays) was compiled for a book on
the preaching of sermons was no accident. Collections
of sermons and books on preaching were by far the
most numerous among the indexed incunabula seen.
At that time sermons were in great demand by local
clerics and preachers who searched these indexes for
suitable themes, especially those appropriate for cer
tain holidays and fast days. Next in popularity were
indexes to theological treatises and philosophical
works which were indexed by the problems that had
been
discussed by philosophers since Aristotle.
Historical works, which enjoyed great popularity
toward the end of the 15th century, were indexed by
the names of biblical personalities, saints, emperors,
kings and heroes as well as by the events in their lives.
Contrary to expectations, very few works dealing with
medicine and the natural sciences were indexed: only
16 out of the 258 incunabula held by the National
Library of Medicine and six items out of 340 in the
Dibner Library (which holds only works on science
and technology) had indexes. A notable exception is
the Gart der Gesundtheit [34], one of the earliest print
ed herbals which has both a classified subject index of
diseases and ailments that could be cured by herbs
(the first index in German) and a bilingual (GermanLatin) name index of plants. A French herbal, the
Arbolaire [4], also had such indexes, modelled on those
of the Gart, as was the case for its translations into
Dutch, Italian and English in the 16th century:
(Wellisch 1978, pp. 84-88). Table 1 provides a detailed
breakdown by subject.
Twenty indexes were found in books printed in the
decade from 1470-9, 19 in those of the decade 1480-9,
and 38 in the last eleven years of the incunabula peri
od. This seems to indicate that, as readers increasingly
came to appreciate the value of indexes, more finding
aids were provided by printers who quickly realized
what Schoeffer had already understood in 1470, name
ly, that the provision of indexes helped to sell books.
Indexes also became generally more elaborate and
voluminous towards the end of the 15th century, a
trend which continued well into the 16th century and
beyond.
TABLE I. Indexes by subject of books
Rank
1
2
3
4
4
4
5
6
6
7
7
Subject
Printed
Manuscript
24
12
10
7
5
7
6
4
2
1
2
2
-
Law
1
-
Total
79
Sermons
Theology
Philosophy
History
Bible commentaries
Medicine
Lives of saints
Biography
Natural science
Literature
Total
%
24
29
12
15
10
12
7
8
7
8
7
8
6
7
4
5
4
5
II
II
83
99
Position of indexes
While tables of contents,
'Tabula', were always found
most often entitled
at the beginning of
books, indexes were either printed immediately follow
ing them or at the end of the text, followed by the
colophon. The position of an index varied even in
cases where the same .index was reprinted in successive
editions of the text, e.g. in [44]. Altogether 56 indexes
could be inspected for their position, 29 of which were
printed in the front matter of the book, while 27 were
printed in the back, a position which became more
common in the 16th century and has been the normal
one since the 17th century.
Terms
Chronological distribution of indexes
The earliest incunabulum index [8] was probably
printed in 1464 or no later than in February 1485 and
not, as has been thought previously, in 1467.
(Householder 1943). Since my description of this index
(Wellisch 1986), a dedication found in a copy of this
work held by the library of the University of Freiburg
im Breisgau indicates that it had been given to its
Faculty of Arts on 14 February 1465 by a certain
Johann Graff who died in 1469 (Sack 1985). The index
is specifically noted in Peter Schoeffer's sales catalogue
of 1470. The first dated index is the one to Speculum
vitae [53], printed in 1468.
The term 'index', found in classical Latin, though
with somewhat different meanings (a person pointing
out something; a list of titles; a title slip; etc.), was sel
dom if ever used in incunabula and became the com
monly used term only in the 17th century. The pre
ferred term was Tabula'; 34 of 56 indexes, or 61%,
had that title. Next in popularity was 'Registrum',
another ambiguous term, because it was also used for
a list of gatherings at the end of books. In its English
form, that term occurs already in Chaucer's 'Knightes
tale' in the sense of a listing:
1 am no divynistre:
"Of soules" find I nought in this registre
The Indexer Vol. 19 No. 1 April 1994
INCUNABULA INDEXES
The term is still in use in German-speaking coun
tries although lately the term 'index' has also gained
acceptance
there.
The
terms
'Repertorium'
or
'Repertorium sive tabula', 'Annotatio', 'Tabula remissoria' (i.e. Table of references) and 'Directorium' were
occasionally also used. Four indexes had no title at all
or were only preceded by an introductory note.
Prefaces and introductory notes
About a third of the indexes seen had either pref
aces, some of which were fairly extensive or they had
at least a brief introductory note about the manner in
which the index had been constructed and how it
should be used. Since there was little uniformity in the
structure of indexes, the nature of items to which loca
tors referred, or the style of locators, it was indeed
necessary to tell readers how to use each and every
index, even though at least towards the end of the
period readers must have become more familiar with
indexes.
The anonymous indexer of the first printed index [8]
wrote a long preface in which he pointed out that
The index and figures of this book are indeed
alone worth its whole price, because they make it
much easier to use... and that Everybody should
also know that the alphabetic letters... set out in
the inner margins serve the said index of the book
which refers to the very same letters... so that
everybody who wants to find quickly something
that is contained in this little book can find it,
and not least also by means of various and many
cross-references it will be revealed what is some
times contained in the diverse passages of this
little book at those points, which will prove to be
most fruitful for those who wish to study the
book.
The compiler of two early indexes, Thomas
Dorniberg, gave a lengthy example for the use and
usefulness of his index to [62]:
If we take the subject of any chapter ... from its
title, a beautiful example is immediately found ...
For instance, from the letter A we take the entry
'Acceptation of gifts leads to four bad effects', the
first of which is that justice is traded for a petty
price. This is treated in Tract ii, chapter 5, section
vii.
The index to Burleigh's De vita et
philosophorum [22] has an introductory note:
moribus
The present little work's very useful table,
arranged
alphabetically
and
containing
the
philosopher's names and their most effective say
ings, begins.
To make sure that no reader would miss the point,
the very same text is repeated after the last index entry
but concluding with the word 'ends'. Typical for
explanations of the structure of entries are the follow
ing:
The Indexer Vol. 19 No. 1 April 1994
In this table, every word is preceded by two num
bers, the first of which indicates the book, the sec
ond the chapter [10].
In one of the many collections of sermons by the
famous preacher Caracciolus [24] the index concludes
with
End of the index of the sermons and of the sub
jects of this book which have been collected in
brief from those sermons, so that everyone who
intends to read about them can easily find a spe
cific sentence wherever it may be, whether in a
sermon to the people or in one to the clerics...
Therefore, he who would like to find anything in
these sermons may look in the index in alphabeti
cal order... And thus it has been provided for
everything for all sentences, references, authorities
and titles which have been indicated in this refer
ence index. Finished successfully.
While such instructions may have been needed in
1473 when only few books had indexes, twenty years
later it was apparently still necessary to state:
Therefore you need to understand that in the fol
lowing index the number described represents the
number of the chapter [46].
Four indexes to German books were found, the ear
liest of which is in the Gart der Gesundtheit [34];
Here follows the fifth and last part of this book,
and it is an index, to find quickly all diseases of
human beings... In this index are also listed the
masters... Avi, that is Avicenna, Ga or Gali,
that is Galienus... Pau, Paulus, etc. and always a
number next to it, namely, i, ii, iii, iv, v, vi, etc.
Other German indexes were found in [23], [45] and
[55].
Types and structure of indexes
Subject indexes were the most common type; 42 of
these were devoted only to subjects, while 13 were
mixed subject and name indexes; 12 books had name
indexes, listing only persons and sometimes places; 3
books had separate name and subject indexes.
Alphabetization
Although virtually all indexes emphasized the fact
that they were 'alphabetically arranged' or 'in the
order of the alphabet', 12 of those that could be
checked for the manner of their arrangement did not
carry alphabetization beyond the first letter of a head
ing; 26 indexes carried alphabetization only to the first
syllable of a heading, that is, to its first two or three
letters, and in only two indexes an effort was made to
file entries also by the letter following the first syllable.
This does not seem to have bothered users too much,
perhaps because in most indexes only a few dozen
entries or less were printed on a page, generally in the
same type size as the text, so that the scanning of
INCUNABULA INDEXES
entries was not too difficult. In five indexes, however,
filing was by the first significant keyword, including its
grammatical variants, for example:
Honor soli deo debetur
Propter honorem suum ...
Contra honorem suum ...
Honor dei includit...
Honor laus et gloria ...
In one instance, subheadings were used in keyword
arrangement [21]:
Charitas describitur. ..
_ et laus eius ponitur . ..
De hac charitate ...
Locators
Foliation and even pagination, with both Roman
and Arabic numerals, was known and used in some
manuscripts of the late Middle Ages (Rouse 1976), but
it was not widely used by the early printers who relied
rather on signatures to indicate the correct sequence of
gatherings for the binders. Indexers continued there
fore to use chapters, paragraphs and sections as well
as marginal letters and a bewildering array of other
means to show at least approximately where a reader
should look for a name or subject. Only towards the
end of the 15th century did foliation and pagination
become more commonly used in books and in their
indexes. The earlier methods had, however, the advan
tage that an index could be compiled before the text
was printed, and it also made indexes 'portable1, that
is, they could easily be reprinted by the same as well
as by other printers, independent of the way in which
the text had been set. This possibility was indeed wide
ly exploited, beginning with the first printed index,
produced by Schoeffer, which was copied by Mentelin
in Strassburg, though in mutilated form (Wellisch
1986). But reprinting of indexes also had disadvan
tages other than deliberate changes or abbreviations
made by subsequent printers, such as the perpetuation
of printer's errors. Thus, the first printing of an index
to Albertus Magnus's work on theology [2] had as its
first entry.
Absolvere quis possit libro vi. ca. xxv ultra medium
The printer of two subsequent editions decided to
number the 'books' and chapters by Arabic numbers
and printed the same entry as
Absolvere quis possit libro 9. cap. 25 ultra medium
Evidently, the typesetter by mistake inverted the
numeral 6 so that the number for the book became 9,
while the chapter number and the indication of the
reference on the page ('beyond the middle') were cor
rectly copied.
Appendix I shows to which extent indexes were
reprinted.
As shown by the example above, most early indexes
used only chapter and paragraph numbers as locators,
often augmented by a series of letters printed at fixed
intervals in the margins in order to pinpoint the place
of a reference on the page, or by adding a verbal
instruction (as above), or both, e.g.,
Abstinentia est multiplex xlv D et infra
Pagination was first used in 1470 by G. Lauer in
Rome, but was only seldom used thereafter by other
printers; only 7 indexes used page numbers as locators,
while most other indexes employed Roman-numbered
foliation numbers. This forced them to resort to cum
bersome means in order to indicate the recto or verso
of a leaf. Marginal letters starting on the recto and
continuing on the verso were one way to locate a
word or phrase, e.g., xii B meant the second para
graph on the recto, while xii M was somewhere on the
verso.
Another method consisted of putting periods either
before or after a folio number, e.g., in [54] which has
Arabic foliation:
Abacus propheta minor .15
Abbacus martyr 32.
Abdias propheta .11 12.
indicating that Abacus (i.e. Habakkuk) is mentioned
on folio 15 recto, Abbacus on folio 32 verso and
Abdias on folio 11 recto and 12 verso. One is left to
wonder why the simple step from folio to page num
bering was not taken as late as 1480, but ingrained
habits are slow to change.
Arabic pagination was sometimes added to printed
Roman foliation by users in handwriting which seems
to indicate that at least some readers were more prac
tically minded than the printers and indexers.
Other locators consisted of column numbers, the
numbers of sermons or philosophical problems in col
lections, books and verses of the Bible, and variously
named combinations of sections, e.g.,
Abscondit se deus a multis. ser. 72, conside. 3,
conclu. 5
which refers respectively to numbered sermons, con
siderations and conclusions.
Cross-references
Only a dozen indexes had more or less sophisticated
cross-references, but among these was [53], the earliest
dated index, printed in 1468, which has, for example:
Artes mechanicae, infra in verbo mechanica
Ludi quae sint liciti aut illiciti, vide infra in verbo
theatrica
In [24] the phrase 'ubi supra' is used when two adja
cent entries refer to the same sermon. The index to
Boethius' De consolatione philosophiae [19] offers terse
'see' references, e.g..
The Indexer Vol. 19 No. 1 April 1994
INCUNABULA INDEXES
Lumen, require tenebrae
Mundus, require deus
Scire, require finis
The last cross-reference refers to the idea that no
one can know his end, and is a good example for the
fact that index users at that time were expected to
know the text of a book almost by heart, especially if
it was a classic.
Double entries were also used, e.g. in the index to
[8], St. Augustine's popular tract De arte praedicandi.
In this, the first printed index, the following entries
occur:
Dicendi modus accendit
Dicendi modus delectat
Dicendi modus flectit
Modus dicendi accendit,
supra, dicendi modus
delectat,
flectit...
et
Since many of St. Augustine's sermons and tracts
were indexed in late 14th century manuscripts, it
stands to reason that this index (which had also other
sophisticated finding aids, including a graphic display
of entries and their relationships) was inspired by if
not copied from earlier manuscript indexes.
In the index of another work by St. Augustine, De
trinitate [13], double entries are also frequently used,
e.g.,
Deus non est conversus in hominem, etc. libro i,
cap, vii
Homo non est mutatus in deum, etc. libro, i, cap,
vii
While this feature made the index easier to use, its
alphabetization was rather crude, by first letter of the
entry only, and even the same entry word was not
always arranged in a single sequence!
Extent of indexes
Only 55 indexes could be inspected for full size,
partly because some indexes lacked one or more of
their first or last pages (probably because they had
been heavily used) and partly because the indication of
indexes in bibliographic sources did not always make
it possible to ascertain the total number of leaves or
pages, much less the number of entries per page or
lines per entry.
The size of indexes varied, of course, according to
several factors, all of which are still valid today, such
as the number of indexable items, the skill of the
indexer, the style of indexing, the size of pages, and
the number of pages allotted to an index by a printer.
Although the number of pages is therefore only a very
rough and ready measure of the extent and compre
hensiveness of an index, yet a general trend towards
longer and therefore more comprehensive indexes is
discernible. The average number of pages for indexes
compiled between 1465 and 1479 is 14.5, for those of
1480-9 it is 18 and those compiled in the last decade
The Indexer Vol. 19 No. 1 April 1994
of the period have an average of 15 and a mode of 20
pages. The largest index encountered (or rather a con
cordance) was one of the handwritten ones with
approximately 5500 items, accompanying a printed
index in [50]. The most extensive among the printed
indexes seen was [32] with more than 2800 entries on
20 leaves for a text comprising only 180 leaves, or 7.8
entries per page; each entry occupies from two to five
lines. One of the earliest indexes, namely that of [27],
also had more than 2000 entries on 24 large leaves for
a text on 266 leaves, or almost four entries per page,
each of which was on two to three lines and some
times more. It was not possible to check a manuscript
of that work (originally written in the first half of the
14th century) for the presence of an earlier index, but
the elaborate nature and the sheer size of the printed
index of medicinal plants with references to their treat
ment in the works of ancient authorities (Dioscorides,
Galen, Avicenna and others) by chapter and section
make it seem likely that it was either copied together
with the text of a manuscript or was modelled on an
earlier handwritten index. The copy seen in the
National Library of Medicine, a huge volume measur
ing 396 x 278 mm, has little index tabs made of parch
ment pasted onto the margins of the first leaves of
each chapter, another quick finding device ordered
from the binder by one of the early owners of the
book.
Taking into account only those indexes that had
more than a thousand entries, the number of entries
per page of text varied from a low of 1.2 to a high of
17, with an average of 5.7, a number comparing favor
ably with modern indexes, although the quality of
those entries is quite a different matter.
Quality of indexes
It has been pointed out by Witty (1965) and
Rabnett (1982) that some incunabula indexes were lit
tle more than listings of proper names and nouns,
compiled by simple underlining of words on proof
sheets which were then alphabetized only by first letter
or at best by first syllable. But these critics examined
only a very small number of indexes, while the sample
investigated here shows rather that, then as now, there
were good and bad indexes and some that were
mediocre. First of all, it goes without saying that mod
ern standards of indexing quality cannot be applied to
indexes compiled in the 15th century (a bias from
which Rabnett is not entirely free). Secondly, although
at least some of the early indexers may have modelled
their work on examples of manuscript indexes, they
had to invent and to improvise as they went along.
But they worked in relative isolation from each other
and became only slowly aware of newer and better
techniques invented in places that were often far dis
tant from their own. Thirdly, what strikes us as flaws
and mistakes, such as only partial or haphazard alpha-
INCUNABULA INDEXES
betization or in the arbitrary omission of names and
topics may not have been perceived as such by the
readers of that time. As Pollard (1908) has pointed
out, indexes were intended to be read and remembered
almost to the same extent as the text indexes. Thus, to
cite just one example, the entry 'Adam et eva quot
annis vixerunt' in Schedel's Liber chronicarum [58] has
no corresponding entry under Eva; but the indexer
may have thought that Eva would always be sought
under Adam since the phrase 'Adam et eva' was then
as now a cliche. Double entries do not occur in the
same work, which shows that the indexer did not
always choose just the first word of an entry as index
term. The criticism of its other faults, pointed out by
Witty, such as foliation as locators without indication
of recto or verso, the lack of cross-references and the
absence of entries for the many illustrations is, how
ever, justified. But this index to one of the most
famous incunabula is not really typical, and its short
comings may have been caused in part by its hasty
compilation; it had to be linked, similar to modern
indexes, to folio numbers and could therefore not be
prepared in advance, as was the case when chapters
and paragraphs were still used as locators. The printer
of this technically complex work was eager to get it
published so as to recoup the considerable amount
invested in the venture, and the indexer may therefore
have worked under heavy pressure, not producing an
exemplary index (Wilson 1976).
Be that as it may, the previous sections have
shown that there were wide variations in the quality of
indexes. Many had only single term entries followed
by chapter or folio numbers, e.g., in [16]:
abissus c.
ableston cxxv
abies cxxxvi
Some indexes, especially those of historical works,
elaborated somewhat, e.g., in [33] (note that Arabic
numerals were used for foliation):
Aadias sextus lydorum rex 50
Aaron primus pontifex 16
Abraam pater 8
Abraam propheta 45
About half of all indexes seen had entries in more
or less elaborate narrative form, sometimes occupying
several lines, e.g., in [46] and [47]:
Absolvere non possunt fratres mendicantes in
casibus pro episcopos reservatos ca. iiij, iii
Misericordiae beatitudo: differt a liberalitate et
magnificentia atque respondet donum consilii:
et pellit avaritiam xxix F
The compilers of indexes were in most instances
eager to point out in a preface that their work was of
high quality, 'very useful' or easy to use. In one case,
the indexer seems to have been convinced that he had
managed to compile the perfect index to Vincent of
Beauvais's Speculum historiale [62]; in his preface he
stated:
May the reader not think that anything is missing
in this Speculum which has not been listed in the
indexes.
On the other hand, the indexer of a German trans
lation of Rolewink's Fasciculus temporum [55] was
apparently not quite sure whether he had listed all
locators correctly:
When you look in the index according to the
order of the ABC, you will find under each num
ber of a leaf where a subject appears. If, however,
it should not appear on that leaf and you are in
doubt, you will find it on another leaf, either
before or after that leaf.
On the whole, it is possible to discern a trend
towards more and better indexes as the 15th century is
nearing its end, thus preparing the way'for indexes of
high quality that were compiled in the 16th century.
The indexers
Who were the compilers of indexes to the sermons,
lives of saints, biographies of lesser mortals, histories
of the world, and books on natural history, medicine,
astronomy and many other topics which the early buy
ers of printed books wanted to find quickly and easily
among the closely printed pages of often huge and
heavy volumes? Some early printers were probably
often also their own indexers, but most of them
employed scholars as editors and correctors whose
tasks included the compilation of indexes. While the
printers had to have, first of all, some considerable
technical knowledge and mechanical skills, most of
them were by no means mere artisans. Many were
learned men and a few even held academic degrees.
Thus, Peter Schoeffer, Gutenberg's assistant and one
of his successors, is known to have been at the
Sorbonne in Paris where he may have earned a law
degree because he became a judge in Mainz and was
also known as a 'clericus'. One of his competitors in
Strassburg, Heinrich Eggestein, had the title 'magister',
and it is known that several other early printers had
bachelor's and master's degrees.
In the first two or three decades of printing, the
printers had to make judicious choices among existing
and well-known works previously available only as
manuscripts, and they tried to anticipate which of
these would find a ready market among the learned
readers. They as well as their typesetters also had to
have a thorough knowledge of Latin, and later on also
Greek, in addition to the various vernaculars which
rapidly became the language of popular (and therefore
profitable) books for a wider audience (Hirsch 1967).
They were assisted in this and other tasks, particularly
The Indexer Vol. 19 No. 1 April 1994
INCUNABULA INDEXES
editing, proofreading and correcting by men of the
highest erudition who were either called upon to per
form the task of editing a particular work or were in
the permanent employ of a printer.
An example is Bernardo Machiavelli (the father of
Niccolo), a prosperous Florentine lawyer, who in 1475
was asked to index Livy's Decades (The history of
Rome) for Nicolaus Laurentii, a printer in his city, as
a spare time job (Eisenberg 1970, p. 100). It took
Machiavelli nine months to index the names of the
cities, provinces, islands, mountains and rivers men
tioned in Livy's work, which seems to indicate that he
its locators refer to folio numbers which have also
been added by the hand of the owner-indexer.
Named indexers
Then as now, the compilers of indexes were seldom
named. In some cases it is obvious from the nature of
the work that the author himself was also the indexer,
especially when the prefatory material to the index is
written in the first person singular. The indexers whose
names have come down to us either cite their monastic
order or indicate that they are learned men. An index
to St. Augustine's De civitate dei [9] by the Dominican
either had only little spare time or was a slow indexer.
friar Nicolaus Triveth was published as early as 1468.
One must remember, however, that an indexer in those
times had not only to do the intellectual job but had
also to make his own indexing slips from scrap paper
as well as sorting boards or wooden boxes (shoeboxes
not having been invented yet). These arduous mental
as well as physical tasks involved in indexing were
described in minute detail some seventy years later by
Conrad Gessner in his instructions for indexers
(Wellisch 1981; Serrai 1989, pp. 48-58).
The early indexers continued a long tradition of
indexing during the last two centuries of the manu
script era, performed by scholars who compiled indexes
for their own use. Sometimes such indexes were added
to those texts that were in great demand and were
produced by scribes in multiple copies. Even a cursory
search in several bibliographies of manuscripts such as
the U.S. and Canadian census (Ricci 1935-40) and the
catalogs of Yale (Shailor 1984) and WolfenbQttel
(Helmsttidter Handschriften 1963) revealed a fairly
large number of indexes in manuscripts that were sub
sequently published by the early printers, particularly
for works by St. Augustine, Boethius, Jacobus de
Voragine and others.
As briefly noted earlier, the sample of indexes con
tained four handwritten ones. The earliest is in one of
the
Folger
Shakespeare
Library's
copies
of
Marchesini's Mammotrectus [50], a dictionary of diffi
cult words in the Bible, explained for the benefit of
clerics and parish priests whose knowledge of Latin
was
rudimentary.
The
book,
originally
named
Mammothreptus (Suckling baby) and first printed by
Schoeffer in 1470, became a bestseller and was reprinted
several times until 1500 and even long thereafter. An
early owner of the Folger copy added an index of the
difficult words, written in a fine gothic hand on eleven
pages inserted after the table of contents; each page
has some 250 entries in three columns, a total of about
2700 entries. Another manuscript index is in the
Folger's copy of Holkot's Postilla super librum
Sapientiae Salomonis of 1479 [40], which is even larger:
it lists names and topics on 24 large pages, each of
which has about 150 entries, a total of some 3600
entries—this in addition to a printed index of about
2200 entries! The manuscript index is dated 1480, and
A monk by the name of Astesanus (or Astaxanus)
compiled an extensive index to his own Summa de
The Indexer Vol. 19 No. 1 April 1994
casibus conscientiae [7] and stated in the preface to it:
I, friar Astaxanus, the compiler of this Summa...
wish to provide this index ... so that whatever
may be required can be found in it with great
ease.
This index was reprinted unchanged eight times by
various printers until 1482. Another Dominican,
Thomas Dorniberg of Memmingen, a doctor of canon
law, compiled a 'Tabula remissoria' (reference index)
to De quattuor virtutibus by Henricus Ariminensis [37]
in 1472 and in the following year, moved 'by the
prayers of many studious clerks' he produced the
index to Albertus Magnus' Compendium theologicae
veritatis [2] which earned him even a mention on the
title page of the book. Also in 1473, Friar Nicolaus
Cerseth of the Dominican Preachers made an index to
Schoeffer's edition of De civitate dei [10]. The indexes
to Schedel's Liber chronicarum and its German edition,
already mentioned earlier, do not reveal the name of
their compiler, but it is possible that either Georg Alt,
the city scribe of Nuremberg, or Peter Danhauser, a
lawyer who is named in the contract between the
printers and investors in this massive project, was the
indexer of both editions (Wilson 1976, pp. 69-70, 243).
Finally, the index to [53] was compiled by a certain
Egidius Daurigni of Beauvais. Thus, in only a few of
the indexes seen were the names of their compilers
revealed, but that is not different from modern prac
tices.
Advertisements
The importance of tables of contents and indexes
for readers of books was soon recognized by the early
printers. Their advertisements and sale catalogs show
that they were vying with each other in making their
books more attractive to prospective buyers by empha
sizing the editorial and typographical qualities as well
as the presence of Finding aids—tables of contents and
indexes.
The earliest advertisement for a printed book that
has come down to us is a sheet printed by Heinrich
INCUNABULA INDEXES
Eggestein in
Bible
Strassburg who in
1466 announced a
... not produced by the art of the pen but by the
marvellous invention of casting letters and print
ing. It has been truly collated with the best texts
by the most excellent men steeped in the human
istic arts... and is in every respect compiled in
the best manner.
While Eggestein advertised the qualities of only one
book, Schoeffer produced a few years later the first
known advertisement of his entire stock of 19 books.
One of these, St. Augustine's De arte praedicandi [8]
was specially singled out as being 'with a notable
table, very useful for preachers'. He did that because
Johann Mentelin, a competitor in Strassburg, had also
printed the same work, but without such a 'very use
ful' finding aid.
Schoeffer clearly perceived the value of a good index
when trying to attract customers for his books. As
indicated above, Dorniberg's index [2] was promi
nently advertised on the title page, so as to distinguish
that edition from others that were produced by various
printers of this popular work, and to attract buyers
who would appreciate the additional value of an
index. Many modern indexers would wish that their
work be acknowledged in such a manner!
Summary
This study is based on a relatively substantial but
biased sample. The bias stems from the fact that about
60% of the sample consisted of medical works and
books on natural science which were deliberately
chosen because my previous study of herbals had
shown that several of the incunabula among them had
indexes (Wellisch 1978). The assumption was that
othef incunabula on those subjects would also have
indexes since they dealt with large numbers of topics
and details such as the names of plants, animals,
minerals, diseases, stars, and techniques which would
be difficult to remember and which would therefore
need finding aids. This, however, turned out not to be
the case: only a small number—6.2% of medical books
and 1.8% of works on natural science—had indexes. If
only the theological, historical and other nonscienlific
book are taken into consideration, the percentage of
indexed incunabula rises to 17%, a not insignificant
amount, given the fact that a large part of the
incunabula such as Bibles, prayer books, indulgences,
romances, proclamations and edicts of the many war
ring temporal and spiritual rulers, and other ephemera
did not need any indexes.
But quantitative measures alone would in any case
be insufficient to judge the state of the art of early
printed indexes. More reliable quantitative and quali
tative data may perhaps in the future become available
when the computerized database of incunabula cur
rently being assembled by British and German
10
researchers will make it possible to search for details
(including indexes) that are now either not or only
insufficiently recorded, buried deep in the printed
bibliographies, and thus not readily retrievable.
The picture that emerges from this admittedly
imperfect study of index production during the last
three and a half decades of the incunabula period
shows that, on the whole, rather remarkable progress
was being made in the provision of finding aids. From
relatively modest beginnings, struggling to invent both
intellectual and technical methods for the compilation
of indexes and their printing, gradually abandoning
chapters and paragraphs as-locators as well as Roman
foliation with its cumbersome numerals and lack of
page indication for the two sides of a leaf, and finally
adopting pagination, the indexers and printers pro
duced increasingly better and more sophisticated
retrieval tools for their readers. What was lost almost
entirely in this process is the early technique of pin
pointing references on a page by marginal letters
which was used by many printers until the 1480s; it
was largely abandoned with the gradual adoption of
pagination and the use of smaller paper formats which
made it easier for readers to find references on densely
printed pages. The handwritten indexes compiled by
owners of books are testimony to the probably widely
felt need for indexes, and the advertisements of
indexes by printers shown to what extent these were
recognized as important and valuable enhancements of
texts. The early printers were in this, as in so many
other
respects,
the
trailblazers
for
the
splendid
books—and their indexes—that were produced by the
printing presses of their heirs in the 16th century and
beyond.
A cknowledgements
I wish to thank the librarians of the Folger Shakespeare
Library, the Dibner Library of the Smithsonian Institution
Libraries, the National Library of Medicine, the Rare Book
Department of the Library of Congress and the Georgetown
University Library who made their incunabula collections
available to me, and enabled me to conduct my study in
many ways.
Appendix 1
INDEXED INCUNABULA
seen or described in bibliographic sources
Note: C = Copinger; H = Hain; HC = Hain-Copinger;
GW- Gesamtkatalog der Wiegendrucke.
1.
2.
Abano, Pietro D\ Expositio problemata Aristotelis.
Mantua: Paulus de Butzbach, 1475. H 16.
Albertus Magnus, Compendium theologicae veritatis
cum tabula Thomae Dorniberg. Speyer: Printer of Gesta
Christi, 1473. GW 597.
Index reprinted Ulm: Johann Zainer, 1478, 1481;
Strassburg: Martin Schott, 1483; Johann PrOss, 1489.
The Indexer Vol. 19 No. 1 April 1994
INCUNABULA INDEXES
3.
6.
Albertus Magnus, De animalibus. Venice: Gregorius de
Gregoriis, 1495. GW 589.
Arbolaire. Besancon: Peter Mettlinger, 1487 or 1488. GW
2312.
Aristoteles. Reperlorium dictorum Aristotelis, Averoys,
aliorumque philosophorum.
Bologna:
Bazalerius
de
Bazaleriis, 1491. GW 2838.
Arnaldus, Stephanus. Commenlum super Nicolaum.
7.
Barcelona: Petro Rosa, 1490. HC 645.
Astesanus. Summa de casibus conscienliae. Strassburg:
4.
5.
Johann Mentelin, not after 1469. GW 2751.
Index reprinted by Mentelin 1473 (twice); Strassburg:
Printer of Henricus Ariminensis, 1474; Basel: Bernhard
Richel, 1477; Venice: Johann of Cologne,
1478;
Cologne: Henrich Quentell, 1479; Venice: Leonhard
Wild, 1480; Nuremberg: Anton Koberger, 1482.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Augustinus, Aurelius. De arte praedicandi. Mainz:
Fust & Schoefler, not after 1465 (1467?). GW 2872.
Index reprinted, Strassburg: Johann Mentelin, not
after 1467.
Augustinus, Aurelius. De civiiate del Strassburg:
Johann Mentelin, ca. 1468. GW 2883.
Augustinus, Aurelius. De civiiale dei. Mainz: Peter
Schoefler, 1473. GW 2884.
Augustinus, Aurelius. De civitate dei. Basel: Johann
Amerbach, 1490. GW 2888.
Augustinus, Aurelius. De sermo domini in monte
habito. Paris: Ulrich Gering & Bertold Remboldt, 1494,
GW 2919.
Augustinus, Aurelius. De trinitate. Venice: Paganinus
de Paganinis, 1489. GW 2927.
Aurifaber,
Aegidius.
Speculum
exemplorum.
Strassburg: Georg Husner, 1487. HC 14917.
Index reprinted by Husner, 1490.
AviCENNA. Canon medicinae. Padua: Johann Herbort,
1476 GW 3116.
Index reprinted by Herbort, 1479; Venice: Peter
Maufer, 1483, 1486; Octavianus Scotus 1490, Simon
Bevilacqua 1500.
Bartholomaeus Anglicus. De proprietatibus rerum.
Basel: Berthold Ruppel, 1470. GW 3402.
Boethius. Opera. Joannes & Gregorius de Gregoriis,
1497/98. GW 4512.
Boethius. De consolatione philosophiae. Nuremberg:
Anton Koberger, 1483. GW 4533.
Boethius.
De
consolatione
philosophiae.
Cologne:
Heinrich Quentell, 1497. GW 4563.
Breitenbach, Johannes de. Repetitio capituli: sententiam
sanguinis Ne clerici vet monachi. Leipzig: Melchior Lotter,
1499. GW 5093.
Brulefer, Stephanus. Opuscula theologica. Paris: Andr6
Brocart, 1500. GW 5587.
Burlaeus, Gualtherius. (Walter Burleigh). De vita et
moribus philosophonim. Cologne: Ulrich Zell, 1470. GW
5781.
Index reprinted Nuremberg: Anton Koberger, 1473,
1477; Friedrich Creussner, 1479; Louvain: Johann v.
Paderborn, 1479/82; Toulouse: Johann Parix, ca. 1480;
Speyer: Johann & Conrad Hist, 1483. Cologne: Johann
Koelhoff senior, ca. I486.
23. Burlaeus, Gualtherius. Das buch von dem lehen und
sitten der heydnischen maister. Augsburg: Anton Sorg,
1490. GW 5793.
The Indexer Vol. 19 No. 1 April 1994
24. Caracciolus, Robertus. Quadragesimale de poenitentia.
Strassburg: Printer of Henricus Ariminensis, ca. 1473.
GW 6066.
Index reprinted Basel: Bernhard Richel, 1475; Basel:
Berthold Ruppel, 1479; Strassburg: Georg Husner,
1479; Strassburg: Martin Schott, 1485; Strassburg:
Johann Grilninger, 1497.
25. Cassiodorus, Magnus Aurelius. Historia ecclesiastica
tripartita. Paris: Georg Wolf, 1492. GW 6166.
26. Dioscorides. [De materia medica]. (Greek). Venice:
Aldus Manutius, 1499. GW 8436.
27. Dondi, Jacopo de'. Aggregator, sive De medicinis simplicibus. Strassburg: Adolf Rusch, ca. 1470. GW 9042.
Chronkon.
Venice: Erhard
28. Eusebius Caesariensjs.
Ratdolt, 1483. GW 9433.
29. Eyb, Albertus de. Margarita poetica. Nuremberg:
Johann Sensenschmidt. ca. 1472. GW 9529.
Index reprinted Rome: Ulrich Han. 1475; Strassburg:
Georg Husner, 1479; Rome: Stephan Plannck, 1480;
Strassburg: Printer of Vitae Patrum, 1483/84, 1485;
Venice: Theodoras de Ragazonibus?, 1487; Venice:
Johannes Rubeus, 1493; Basel: Johann Amerbach,
1495.
30. Ficinus, Marsilius. De triplici vita. Basel: Johann
Amerbach, 1497. GW 9885.
Index reprinted Venice: Bartholomeus Pelusius, 1498.
31. Folz, Hans. Von den heissen Bttdem. Nuremberg: Hans
Mair, 1495. GW 10115.
Index reprinted Brno: Clement von Gracz, 1495.
32. Foresti, Jacobus Philippus Bergomensis. Supplementuni chronicarum. Brescia: Boninus de Boninis, 1485.
HC 2806.
33. Foresti, Jacobus Philippus Bergomensis. Supplementum chronicarum. Venice: Bernardinus Rizus, 1483.
HC 2809. Reprinted by Rizus 1492/93.
34. Gart der Gesundtheit. Mainz: Peter Schoeffer, 1485. H
8948.
35. Gesta romanorum. Louvain: Johannes de Westfalia, ca.
1484. C 2716.
36. Guilelmus Arvernus. De fide et legibus. Augsburg:
Gunther Zainer, 1475. H 8317.
37. Henricus Ariminensis. De quattuor virtutibus cardinalibus. Strassburg: Printer of Henricus Ariminensis, 1472.
H 1649.
Index reprinted Speyer: Printer of Gesta Christi, 1472.
38. Herpf, Henricus de. Speculum aureum decem praeceptonan dei. Mainz: Peter Schoeffer, 1474. HC 8523.
39. Hieronymus. De viris illustribus. Strassburg: s.n., 1483,
1485. HC 8600.
40. Holkot,
Robertus.
Super
sapientiam
Salomonis.
Cologne: Conrad Winters, 1479. HC 8755.
41. Hortus sanitatis. Mainz: Jacobus Meydenbach, 1491. HC
8944.
42. Iacobus de Voragine. Dominicales per totum annum.
[Germany], s.n., 1484. C 6534.
43. Jacobus de Voragine. Legenda aurea sanctorum.
Cologne: Conrad Winters, 1476. C 6410.
44. Jacobus de Voragine. Legenda aurea sanctorum.
Strassburg: Georg Husner, 1492. C 6458.
45. Jacobus de Voragine. Leben der Heiligen. Nuremberg:
Anton Koberger, 1488. HC 9981.
46. Lochmaier,
Michael.
Parochiale
curatorum.
Nuremberg: Friedrich Creussner, 1493. HC 10167.
11
INCUNABULA INDEXES
47. Ludovicus Pruthenus. Trilogium animae. Nuremberg:
Anton Koberger, 1498. HC 10315.
48. Maoninus Mediolanensis. Regimen sanitatis. Basel;
Nicolaus Kesler, 1493. H 10486.
49. Maioranis, Franciscus de. Sermones ab adventu cum
quadragesimalis. Venice: Bernadinus Rizus, 1491/92. HC
10530.
50. Marchesinus, Johannes. Mammotrectus super Bibliam.
Strassburg: Printer of Henricus Ariminensis, ca. 1472.
HC. 10552.
Index reprinted Venice: Franciscus Renner, 1478.
51. MOsch, Johannes. Tractatus de horis canonich dicendis.
Augsburg: Anton Sorg, 1489. HC 11534.
52. Repetitiones totius iuris canonici capitulorum. Paris:
Poncetus le Preux, ca. 1500. HC 13876.
53. Rodericus (Sancius) Zamorensis.
Speculum
vitae
humanae. Rome: Sweynheym & Pannartz, 1468. HC 13939.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
Index reprinted Augsburg: Gunther Zainer, 1471;
Paris: Ulrich Gering, 1472.
Rolewinck, Werner. Fasciculus temporum. Venice,
Erhard Ratdolt, 1480. HC 6926.
Index reprinted by same printer 1485.
Rolewinck, Werner. Ein burdlin [sic] der zeyt.
Strassburg: Johann Priiss, 1492. HC 6940.
Rudimentum novitiorum. Liibeck: Lucas Brandis, 1475. H
4996.
Sabundus.
Raymundus
de.
Theologia
naturalis.
Strassburg: Martinus Flach, 1496. HC 14069.
SCHEDEL, Hartmann. Liber chronicarum. Nuremberg:
Anton Koberger, 1493. HC 14508.
Schedel, Hartmann. Das Buck der Chroniken und
Geschichten. Nuremberg: Anton Koberger, 1493. H
14510.
Sylvaticus, Matthaeus. Liber pandectarum medicinae.
Strassburg: Adolf Rusch, ca. 1480. HC 15192.
Vincentius
Bellovacensis.
Opera.
Venice:
H.
Liechtenstein, 1494. C 6241
Vincentius
Bellovacensis.
Speculum
historiale.
Strassburg: Adolf Rusch, ca. 1473. C 6245.
Index reprinted Strassburg: Johann Mentelin, 1473.
Vincentius
Bellovacensis.
Speculum
morale.
Strassburg: Johann Mentelin, 1476. C 6252.
Appendix II
Printing places of incunabula indexes
printed during the first twelve years after the first one
Year
Place
Printer
Work by number in
Appendix I and author
1465? Mainz
SchoefTer
1467 Strassburg Mentelin
1468 Rome
Sweynheym &
Pannartz
1470 Cologne
Zell
Strassburg Rusch
Basel
Ruppel
1472
1473
1474
12
Nuremberg Sensenschmidt
Nuremberg Koberger
Strassburg Printer of Henricus
Ariminensis
Mainz
Schoeffer
8 Augustine
8 Augustine
53 Rodericus
Zamorensis
22 Burlaeus
27 Dondi
16 Bartholomaeus
Anglicus
29Eyb
23 Burlaeus
24 Caracciolus
38 Henricus Herpf
1475
Mantua
Basel
1476 Padua
Cologne
Butzbach
1
24
15
43
Richel
Herbort
Winters
Pietro d'Abano
Caracciolus
Avicenna
Jacobus de Voragine
References
Dachs,
Karl
and
Schmidt,
Wieland.
Wieviele
Inkunabelausgaben gibt es wirklich? Bibliotheksforum
Bayern 2 (1974): 83-95.
Eisenberg, Elisabeth. The printing press as an agent of change.
Cambridge: Cambridge University Presss, 1979, 100.
Die
Helmst&dter
Handschriften.
Frankfurt
a.
M.:
Klostermann,
1963-. (Kataloge der Herzog-AugustBibliothek Wolfenbiittel... Neue Reihe)
Hirsch,
Rudolf.
Printing,
selling and reading
1450-1550.
Wiesbaden, Harrassowitz, 1967.
Householder, Fred W. The first printer, The Library, 4th
series, 24 (1943): 30-46.
Pollard, Alfred W. Indexes. Cornhill Magazine 97 (1908):
232-242.
Rabnett, Mark. The first printed indexes: a study of indexing
techniques
in
some
incunabula.
Cataloging
&
Classification Quarterly 2 (3/4) (1982): 87-102.
Ricci, Seymour de. Census of medieval and renaissance manu
scripts in the United States and Canada. New York: H. W.
Wilson, 1935^0.
Rouse, Richard H. Cistercian aids to study in the thirteenth
century. (In: Studies in medieval Cistercian history II.
Kalamazoo, MI: Cistercian Publications, 1976, pp.
123-134).
Sack, Vera. Die Inkunabeln der UniversMtsbibliothek in
Freiburg im Breisgau... Wiesbaden, Harrassowitz, 1985.
Serrai, Alfredo. Le enciclopedie rinascimentali. 8. Bibliografi
universali. Conrad Gesner. Pandectae. // bibliotecario
(20-21) (1989): 1-105, pp. 48-58.
Shailor, Barbara A. Catalogue of medieval and renaissance
manuscripts in the Beinecke rare book and manuscript
library, Yale University. Binghampton, NY: Renaissance
Texts and Studies, 1984.
Steinberg,
S.
Hr Five
hundred years
of printing.
Harmondsworth (England): Penguin Books, 1955, 19.
Wellisch, Hans H. Early multilingual and multiscript indexes
in herbals. The Indexer 11 (Oct 1978): 81-102.
Wellisch, Hans H. How to make an index—16th century
style:
Conrad
Gessner on indexes and catalogs.
International Classification 8 (1981): 10-15.
Wellisch, Hans H. The oldest printed indexes. The Indexer
(Oct 1986): 73-82.
Wilson, Adrian. The making of the Nuremberg Chronicle.
Amsterdam: N. Israel, 1976.
Witty, Francis J. Early indexing techniques: a study of several
book indexes of the fourteenth, fifteenth and early six
teenth centuries. Library Quarterly 35 (1965): 141-148.
Witty, Francis J. The beginnings of indexing and abstracting:
some notes towards a history of indexing and abstracting
in Antiquity and the Middle Ages. The Indexer 8 (Oct
1973): 193-198.
Hans H. Wellisch is Professor Emeritus at the College
of Library and Information Services of the University of
Maryland, and author of Indexing from A to Z, the
recently published guide and style manual for indexing,
among several other books and articles.
The Indexer Vol. 19 No. 1 April 1994