As of 22 April 2009 this website is 'frozen' in time — see the current IFLA websites

This old website and all of its content will stay on as archive – http://archive.ifla.org

IFLANET home - International Federation of Library 
Associations and InstitutionsAnnual 

64th IFLA Conference Logo

   64th IFLA General Conference
   August 16 - August 21, 1998


Code Number: 103-158(WS)-E
Division Number: IV.
Professional Group: Classification and Indexing
Joint Meeting with: Co-sponsored by UDC/FID/BSI: Workshop "UDC"
Meeting Number: 158.
Simultaneous Interpretation:   No

The Revision Process in UDC; An Examination of the Systematic Auxiliary of 'Point-Of-View' Using Facet-Anaytical Methods

Vanda Broughton
University College London
London, UK


The process of maintenance and revision in a large classification scheme is more than an editorial activity; current thinking about indexing theory needs to be a part of the revision process. On-going work on the UDC employs modern techniques to produce a structurally sound classification based on facet analytical methods; this process is demonstrated using examples of recent revision in the Auxiliary Schedules. Analysis of the 'Point-of-view' table (Schedule 1i) reveals the lack of logical principles underlying this auxiliary, and indicates the need for a new systematic schedule, of properties.


The nature of classification revision

To the interested observer the process of maintaining a major classification scheme may well appear to be a predominantly editorial exercise; new subjects, such as superstring theory or silicon chip technology, arise along with a new vocabulary, and these must be incorporated into the scheme. New events, discoveries, and individuals must similarly be accommodated by any system, their positions decided, notation allocated, the index amended, and all these decisions made known to the user. In the same way older subjects fall from favour, and no longer generating a literature, must be pruned away by the reverse process.

At the same time any serious programme of revision must attempt to do more than this, more than classificatory housekeeping. Attention must be paid to developments not only in the subject matter with which the classification deals, but also in the discipline of subject indexing itself, for the indexer is not concerned exclusively with the material content of the system but also with its effectiveness as a tool in the organization and exploitation of documents; this remains true at whatever level and for whatever purpose an indexing language is used, whether pre-co-ordinately, post-co-ordinately, manually, automatically, for books, artefacts, print or electronic, real or virtual forms, or whatever. The true measure of the viability of a scheme lies in its logical principles, in flexibility of use, hospitality to new ideas, to synthesis, to its capacity for dealing with complex notions. Secondary features will include ease of use, predictability (in the location of compounds), the acceptability of notation, and so on, but all are subordinate to the infrastructure as displayed in the schedule construction and order.

In most modern systems of classification, general or special, the structural principles will reflect those ideas of categorical analysis formulated by Ranganathan and developed subsequently into the full theory of faceted classification. That is, the provision, for any given subject, of a relevant vocabulary, the individual terms of which are analysed and organized into functional categories of the type 'thing - kind -part - property - process - operation - patient - agent - space - time - form', with a given citation order between facets, and rules for compounding both conceptually and notationally.

Facet analysis as applied to the UDC

UDC contains most of the features desirable in a general scheme, particularly the synthetic principle, by which the individual components of a compound subject may all be expressed notationally using the colon. Nevertheless it has not until recently embraced a fully faceted approach to its analysis of subjects. The application of facet methodology to the vocabulary of UDC and the introduction of the structure derived from this method underlies much of the revision work now in hand.

Facet analysis and the auxiliary schedules

Recent work on the auxiliary schedules, Tables 1a-k, includes several examples of this approach in operation. From the list of functional categories given above, the general categories of space, time and form are recognised by the majority of schemes, and systematic schedules, which constitute the facets of Place, Period and Physical Format for combination with any subject, are normally provided; UDC is no exception to this, and its auxiliary schedules for the facets of place, period and physical format are particularly comprehensive and detailed. The type of revision undertaken on the area table Schedule 1e, (for example the recent revision of Physiographic regions in E and C 1997) which largely consists of editorial work as described above, (i.e. updating vocabulary in line with changes in political and administrative boundaries, correcting errors of content and typography, and so on) is typical of maintenance activities in a part of the schedules already consistent with facet principles.

The creation of Schedule 1k (and the revision of the whole classification which accompanied it) represents the logical development of this, in that it confirms the existence of a category, or facet, which has previously only been implicit in UDC, namely that of persons (or agents) in the subject. (The recently introduced auxiliary for the category of materials is a parallel example of the same process.)

In a fully faceted system any given concept is combined with any other concept where the need arises; compound subjects are not distributed throughout the scheme by enumerating them as sub-classes of a containing class; for example, the term 'lawyers' is not a subject subdivision of the class Law, but should be derived by synthesis via the process which analyses it as a compound subject 'law - persons in the subject'. This compound might still be listed under the heading 'Law', but conceptually and notationally it occupies a specific position determined for it by the process of synthesis. (As a consequence, to the characteristics of flexibility and hospitality, are added those of predictability and logical consistency within the overall structure of the scheme.)

It is therefore part of the work of revision to identify where these 'distributed' categories occur and to bring them together as a discrete group, the members of which can be compounded with any other class as required, in a systematic and consistent manner.

Auxiliary Schedule 1i - the 'Point-of-View' table

The concept of a 'point of view'

What, if anything, can be said to constitute a legitimate 'point of view', and can we establish that this is both contained in this schedule, and most usefully expressed by the use of a .00 number?

It is in the nature of a bibliographic classification (which deals with documents), as opposed to a taxonomic one (which deals with objects), that every item to be classified will have a point of view; otherwise every work ever produced on a given subject would be identical in content (or at least similar to the extent that boron molecules or chickens are similar). This is patently not the case, and we should therefore assume that a 'point of view' is inherent in any work of authorship. Whether it is useful to express this point of view in a classmark is another issue.

Sometimes it is undoubtedly the case that the author's viewpoint is significant; for example, the creationist's view of evolution will differ substantially from that of the Darwinian biologist. (Even so it may not be particularly helpful to express this fact in a class number, other than by the simple use of the colon.) However, most of the time the author's viewpoint is irrelevant; for example most science books written by Christians will not display anything other than a scientific attitude, and the author's viewpoint need not even be acknowledged, let alone expressed in the classmark. Nevertheless this is the only way in which the 'point of view' could affect the content of the document, and we need to consider whether it is sufficiently different to warrant treatment other than the straightforward use of the colon. The common auxiliaries of form include provision for specifying the 'audience' for a document at (0.05) e.g. books for children; it would be logical to expand this to include the inverse relationship e.g. fiction by women, poems by children. This approach would also accommodate many conventional 'point of view' documents, particularly those of the type which are often described as 'treatments' (e.g. a humorous treatment, a fictional treatment).

Analysis of the auxiliary schedule for Point of View

When we look at Schedule 1i in detail however, the terms found there do not appear to belong to a 'point-of-view' category in the sense described. It seems that the original compilers of UDC did not use 'point-of-view' in any precise sense, but rather as a portmanteau word for those terms which they wished to apply generally, but the function of which they did not define very closely. It is a kind of 'ur-auxiliary schedule' acknowledging the need for synthesis of concepts other than the categories of space, time, and physical form. It is, in that respect ahead of the time at which it was written, but fails, by modern standards, to offer a sufficient solution to the problem of generally applicable classes.

As it stands the auxiliary table still exhibits the tentative structure of the original table, and the lack of a single coherent underlying principle is quite evident. A detailed analysis of the schedule reveals the existence of several different categories and characteristics of division, which conflict one with another, and which replicate and overlap a number of the other auxiliary schedules ( and indeed some parts of the main classification, synthesis with which could be more economically and elegantly expressed by the use of the colon, rather than by the application of a common auxiliary).

Structure and Content of Table 1i

When using a standard facet analytical approach to the schedule it is clear that several different and mutually exclusive categories are operating simultaneously, and that a proper distinction fails to be made between them. Those most easily identified are:~

  1. Common subdivisions of form. This includes both inner form (that is, a common 'subject' approach such as research, or theory of a subject) and outer, or physical, form (plans, drafts, etc.), for example:

    The intentions of the compiler seem doubly vague as it is not really clear whether we are dealing with say, a document about calculations or a document containing calculations

    1. Terms which belong to a general operations facet (testing, monitoring, production…) Many of these are also represented in the special auxiliaries in Class 6, and can be compounded from that main Class, using the colon for combination.

      • .001.53     Laboratory research
      • .002.71     Transport. Handling
      • .004.68     Adaptation. Alteration for different purposes

      Parts of a general properties facet (equality, compatibility, capacity….) These are otherwise very widely occurring terms, although some aggregations, as at spatial properties in Schedule 1e, or in the inadvertent use of properties as specifiers (or species makers) do occur.

      • .001.361     Similarity. Resemblance. Analogy
      • .002.63       Raw products

    2. A rudimentary persons (or agents) schedule, at .007, which is in direct conflict with the Common Auxiliary of Persons at -05 (workers, experts, etc.)

      • .007.1     Top management         cp.     -057.177 Top management
      • .007.3     Clerks                         cp.     -057.3 Clerical workers

    All of these are quite fragmentary in nature, and none of them are taken to their logical conclusion or worked out in detail. Neither is there any consistency in the grouping of terms within the schedule, so that it is not clear which of a number of options might be used, and related concepts are quite widely separated; for example the idea of 'plans' occurs in four different places:

    or financial aspects:

    or amendments, modifications:

    and so on.

    The overall impression is of a collection of vaguely related terms, rather than a systematically constructed table. Partly this arises from the difficulty in separating the various categories of terms included here (on account of their very general characteristics and nature) but also because of the rudimentary nature of synthetic classification at the time of this schedule's development. The table is highly reminiscent of the later Systematic Auxiliaries of the Bibliographic Classification of H. E. Bliss, who also recognised the need for a pervasive synthesis, but failed to work the principle through, beyond the embryonic structure found in the first edition of his classification.
    If the objective of a fully faceted approach is to be realised, it is necessary to rationalise the content of the Point of View table and to distribute to the other auxiliaries and/or main classes those terms which are clearly not descriptive of a document's standpoint.

    Otherwise the detail of Schedule 1i seems not to deal with relationships between subjects, but with the application of a general operations and properties facet, which is at present not separately represented among the common auxiliaries

    Solutions to the problems presented by Schedule 1I and further development:-

    It can be assumed that for the main part relationships between subjects are adequately dealt with by the compounding devices + and /, and the limiting device of the colon. 'Point-of-view' is used in this schedule as an inaccurate synonym for 'in-conjunction-with', a very generalised relationship which cannot really be said to improve on that provided by the colon.

    Schedule 1i does not provide a unique location for most of its constituent classes; as a result there is potential conflict and confusion as to the preferred treatment of terms located here. Alternative class numbers for nearly all existing Point of View classes can be found elsewhere in the schedules, whether these are alternative locations, new numbers, or classmarks synthesized from existing numbers.

    Those terms identified as belonging to either form or persons categories can be derived from the appropriate auxiliary schedules (the auxiliary for Form containing the only valid point-of-view concept - that of documents written by or for particular classes of persons). Most of the terms which can be categorised as operations or methods can be found in various sections of the technology class 6 - representing as they do, activities within management and production; these can be compounded directly with any classmark without the need to employ Schedule 1i.

    A residue of terms remain which appears to belong to a general category of properties. This suggests that a systematic auxiliary needs to be provided for these, along the lines of those already existing for persons and materials; that is a schedule of general concepts which can be required in a number of places throughout the scheme, and for which a consistent and logical mechanism for synthesis should be provided. It is true that some properties are already isolated - namely those spatial and temporal properties enumerated in the Place and Period auxiliaries, and material properties and characteristics of persons - but a greater range of properties need to be catered for.

    To this end the whole of the scheme is now being analysed for examples of property terms currently subsumed to specific subject areas, but which need to be made more generally applicable. A preliminary survey has indicated that there are indeed a large number of general properties of this kind, and a draft proposal is now in preparation for a new auxiliary schedule.