The titles constituting the catalogue may be variously arranged. They
may be placed under the names of authors, and the names disposed in
alphabetical order; they may be grouped in classes, according to
subjects; or they may be made to follow the order of the date, or
place of printing.
The two most common forms for catalogues, are the alphabetical and
the classed. Much controversy has arisen respecting their comparative
usefulness. It is not necessary to revive it here, since the system
now proposed, renders it easy to vary the order of titles, so as to
suit any desired form.
For the General Catalogue, however, it is, for several reasons,
desirable to adopt the alphabetical arrangement.
It would be impossible to propose any system of classification,
which would command general approval, or upon which a commission
of competent bibliographers would be unanimous in opinion. A
classification, founded upon the nature of things, though it
has occupied the best thoughts of such men as Bacon, Leibnitz,
D'Alembert, Coleridge, Ampère, and many others, has not yet been
attained. Every classification which has been proposed or used, is
more or less arbitrary, and consequently unsatisfactory, and liable
to be altered or superseded.
If, however, it were possible to agree upon a system of
classification, the attempt to carry it out would, in a work like
that proposed, be fatal to uniformity. Where different men were
applying the same system, their opinions would vary, with their
varying intelligence and skill. This would lead to utter and
irremediable confusion, and would eventually defeat all our plans.
Even were these objections obviated, the occurrence of fewer
difficulties in constructing an alphabetical catalogue would still
present a decisive argument in its favor. Even these are great. If
increased, by an attempt at classification, they would soon lead to
an abandonment of the work.
Another consideration of great weight is, that, in reprinting
classified catalogues, and inserting additions, if the titles were
kept in systematic order, the work of selecting those to be used,
and of distributing them to their places, would have to be done
by a person, who, besides being a practical printer, should be
familiar with the bibliographical system adopted. This would be very
expensive. Whereas, on the alphabetical plan, any printer could do
the whole.
On general considerations, without special reference to those which
are peculiar to this system of publishing, alphabetical catalogues
are to be preferred;--catalogues in which all the works of each
author are placed under his name, and the names of authors are
arranged alphabetically; anonymous works being entered under the
first word of the title, not an article or preposition. Such is now
the general opinion of competent bibliographers and literary men.
The Edinburgh Review, in an able and interesting article upon the
British Museum, holds the following language:
"It seems to have been almost universally agreed that the
catalogue ought to be alphabetical. Some time ago the
current of opinion among literary men seemed to be setting
towards classed catalogues, or those in which the books are
arranged according to subjects. We had hardly supposed that
this illusion (as we hold it to be) had become so nearly
obsolete as the evidence before us shows that it is: and
this disappearance of a most injurious opinion, which never
was entertained to any extent by the really experienced
in bibliography, encourages us to hope that it will not
be long before the _professional_ persons just alluded to
[librarians] will be admitted to know best on all the points
which have been raised relative to the care of a large
library."
The experience of all students, of all who use books, if carefully
noted, will show, that, in a vast majority of cases, whoever wishes
to refer to books in a library, knows the names of their authors. It
follows, that this form of arrangement must be, in the main, the most
convenient; and if any other be pursued, it can but accommodate the
minority, at the expense of the majority.
Still, it is indisputable that, oftentimes, the names of authors
are not known; that one knows, merely, what subjects he wishes to
investigate.
It may be said, that a catalogue, being designed to be merely a list
of books contained in a library, is not to be expected to furnish
this information; and that references to all authors, treating of
any particular subjects, may be obtained from bibliographical works,
encyclopædias, and other sources of information. This is true. But,
unfortunately, these sources of information are not generally known,
or not readily accessible, even to men of considerable attainments
and scholarship.
It becomes, then, a question of importance how far the wants of such
persons are to be provided for. The following remarks on this subject
are worthy of attentive consideration:
"On this, as on other points, we may observe that two
descriptions of persons consult a catalogue--those who know
_precisely_ what book they are in search of, and those who
do not. The first will find by any rule, so soon as they
have learnt it; and will be glad indeed of a catalogue which
preserves its consistency, even though 600,000 titles,
running over four quarters of the globe, four centuries of
time, and four hundred varieties of usage, should actually
require _ninety-one_[1] rules of digestion. The second class
could easily be suited, if all their imperfect conceptions
tended to the same case of confusion: and, as being the
majority, would have a right to the adoption of the one
nearly universal misconception; which, being one, would
furnish a rule. But it is truth which is single, while
error is manifold; and consequently, it is clear to every
common sense except that of men of letters claiming, as
such, to be bibliographers, that one of two things should be
done:--either the truth should be taken, when known, or in
the event of it being possible to be wrong, the error should
be the consequence of a digested and easily-apprehended
rule, consistently applied throughout. If the framer of the
catalogue be allowed to do as he likes, the consulter of
it must do as he can. Now which of the two classes should
be considered in preference,--those who know what they
want or those who do not? The Doctor of Divinity already
quoted, gives this as one of his rules: 'Item, whan anie
man comith and wotteth not what he wold haue, then he (the
keper of the Bokys) shall tell hym, and doe hym to understond
hys besynesse.' This can be done, to a certain extent,
by _cross-references_. But, all cross-references being
concessions to want of accurate knowledge, it is plain that
discretionary entries, with discretionary cross-references,
would form a plan which puts entirely out of the question
the convenience of the person who knows exactly what he
wants; which kills both calf and cow for the less deserving
son, without giving the power of making any answer to the
complaint of the one who never fed on husks. Nothing is
stranger in the course of the evidence before us, than the
quiet manner in which the opponents of the existing plan take
it for granted that no one ever goes with a precise knowledge
of the title-page of the work he seeks, unless it be the
coolness with which this accurate inquirer is told, as Mr.
Carlyle said to those who write useful knowledge, that he
is one 'whom it is not worth while to take much trouble to
accommodate.'"
[1] These are not all that might be wanted. For example,
the case is not provided for, though it has occurred, in
which an author, in his title-page, invites the reader to
make his choice between two ways of spelling his own name.
Here, we are to presume, some of our witnesses would take
the first method given, others would leave the cataloguer
to comply with the author's request.
But it is convenient even for those to whom the principles and
means of research are best known, to be able to ascertain, readily,
what books, of those which they know to have been written upon the
subjects of their investigation, are to be found in the particular
libraries which they consult. This end may be attained in the
following manner. In connection with the _catalogue_ of each library,
there should be an _index_ of subjects. This index should also be
alphabetical. Under each subject, the divisions which naturally
belong to it, should be distinctly recognized. It may here be
remarked, that the parts of any particular science, or branch of
learning, may be clearly defined, and universally acknowledged,
whilst the relation of this science, or branch of learning, to
others, may not be clearly established. To use the words of a
vigorous writer upon this subject: "Take a library upon one science,
and it classifies beautifully, sketching out, to a nicety, the
boundaries, which, it is too rarely noticed, are much more distinct
between the parts of a subject, than between one subject and another.
Long after the counties of England and Scotland were well determined,
the debateable land was nothing but a theatre of war."
This index should be alphabetical, rather than classed, because it
is easier to find a word, in an alphabetical arrangement, than in
any other order of classification; and, besides, the subject of
research may be one not admitted, as a distinct division, in any
classification. Such indexes can hardly be expected, immediately, in
connection with the general catalogue; though, it is to be hoped,
that these valuable appendages will not long be, of necessity,
omitted.
A method of securing uniformity in such indexes may, hereafter, be
agreed upon, so that they may be combined and form an alphabetical
index of subjects to the general catalogue. It is thought best,
however, for the present, to limit our efforts to the procuring
of good alphabetical catalogues, as a groundwork, to which other
valuable aids to research, may, as opportunities offer, be superadded.