LANGUAGE AND CLASSIFICATION
| Published date | 01 April 1965 |
| Date | 01 April 1965 |
| DOI | https://doi.org/10.1108/eb026378 |
| Pages | 275-278 |
| Author | D.J. FOSKETT |
| Subject Matter | Information & knowledge management,Library & information science |
LANGUAGE AND CLASSIFICATION
D.J. FOSKETT
Librarian, University of London Institute of Education
ONE OF THE least studied aspects of information retrieval
is
the relation-
ship,
on the linguistic plane, between the language of enquirers and the
terms used by indexing
systems.
Much of the writing about system structure
assumes that these are equivalent, that every enquirer is honest enough to
call a spade a
spade,
and that the problems of designing descriptor languages
for use in indexing he in the field of definition. This partly accounts for the
sudden leap to fame of the thesaurus; for while a dictionary merely gives
definitions of words one by one, a thesaurus, being slightly more refined,
gives descriptive analyses in terms of synonyms and near-synonyms, some-
times grouping them into sets based on some common characteristics.
But the nodal activity of information service is not the translation of
authors'
terms into an indexing language. If it were, how simple our prob-
lems would
be!
We should require no more than
a
sort of conversion table,
which could always be brought up to date by the incorporation of any new
terms used by an author, in a manner consistent with the index structure.
No difficulty here; an author's text
is
a static, completed unit, which gives
to all
its
terms a stated context. The real nub of information
service,
on the
contrary, is to express an ill-defined, incomplete, and changeable thought
pattern—the
enquirer's—in
terms that will bring out of a store information
able to change that pattern into a coherent system corresponding with
reality. Barbara Kyle's article in the
Journal
of
Documentation
last year in-
cludes a graphic illustration of the process, and underlines its importance.1
A detailed analysis
is
long overdue, but the science of language itself can-
not claim to be far advanced; only in recent years have we broken away
from the speculative approach of nineteenth-century comparative phil-
ology (which however laid the foundations) and begun to study language
as a phenomenon sui
generis.
One
thing,
however,
is
certain: the scattering—
indeed the isolation—of terms implied by alphabetical organization is not
'natural', in either of two important
senses,
even though at
first
sight it may
appear as simple as ABC. Alphabetical arrangement of terms corresponds
to nothing. It breaks up knowledge into fragments surrounded by others
which are not only unrelated but also different from one language to an-
other. This
is
not 'natural' because
a,
reality, of which knowledge
is,
or pur-
275
To continue reading
Request your trial