Subject Cataloguing: Chain Procedure, POPSI and Precis

epgp books

Structure of Module: E-Text

 

1. Introduction

2. Chain Procedure

2.1 Definition

2.2 Functions

2.3 Pre-requisites

2.4 Method of Chain Procedure

2.5 Class Index Entries

2.6 Advantages

2.7 Limitations

3. POPSI

3.1 Introduction

3.2 Advantages of POPSI

3.3 Steps in POPSI

3.4 New version of POPSI

3.5 Working of POPSI

3.6 Potentialities of POPSI

4. PRECIS

4.1 Introduction

4.2 Definition of PRECIS

4.3 Chief characteristics of PRECIS

4.4 String Making

4.5 Role Operators

4.6 Generation of Entries

4.7 PRECIS Vs Chain Indexing

5. Questions

6. Further Readings

 

1. Introduction

 

Subject approach is the most important approach of the library clients. This approach is served in dictionary catalogue by preparing subject entries with the help of printed list of subject headings i.e. L.C. List of Subject Headings or Sears‟ List of Subject Headings, which have been discussed in Module 6. In Classified catalogue subject entries are prepared according to Chain Procedure devised by S.R. Ranganathan. In this module Chain procedure, POPSI and PRECIS are discussed.

 

2. Chain Procedure 2.1 Definition

 

The unique concept of Chain Procedure was first expounded by Ranganathan, a staunch supporter of subject approach, in his Theory of library catalogue in 1938. “It is the digit by digit translation of the class number, assigned to a document, into ordinary language”. It means each class number is analyzed as a series of links i.e. steps of division from the main class to the specific subject. Out of this analysis, sought links are used as the headings of class index entries beginning from the last step. Ranganathan has defined it as a “procedure for deriving subject word entry from a class number, in a more or less mechanical way”. Once a person understands the technique of Chain Procedure he can draw it smoothly and almost mechanically. Chain Indexing or Chain Procedure as it is popularly known as defined by Batty C David is “a method of deriving alphabetical subject index entries in a semi automatic fashion from the chain of successive subdivisions that leads from a general level to the most specific level needed to be indexed. It may be used to provide indexes not only to classification schemes and classified catalogues, but also to all other systematically organized indexes, even when they are arranged alphabetically. The chain is nearly always, but not necessarily, taken from a classification scheme in use, and the method is intended to offer general as well as specific access to all information and thus to avoid for arbitrary decisions by the indexer as to how a subject may be approached through the catalogue”. It is a “Semi mechanical method of producing the requisite subject index entries for a classified catalogue based on analysis of the classification symbol of each subject”. E J Coates has defined it as a method, first propounded by Ranganathan, for constructing subject index entries, without permutation of components, by citing terms contained in particular chains.

 

2.2 Functions

 

A Thirmalaimuthuswamy has enumerated the following functions of Chain Procedure:

 

(a) To derive the subject word headings to serve as an unfailing index to the arrangement of subjects in the classified part of the catalogue.

(b) To display many of those relations between the component parts of a specific subject which are not displayed by the classified sequences; and

(c) To provide the required alphabetic subject index key to find out the class number of a particular subject for consulting the classified part.

 

2.3 Pre-requisites

 

The Chain Procedure works successfully and efficiently on the following conditions:

 

(a) The book must be classified first and catalogued later. It means that a book would be read technically by the classifier and not by the cataloguer;

(b) The class number alloted to the book should be co-extensive i e it should represent the specific subject of the book, through a proper befitting class number;

(c)  Classification scheme adopted for classification of documents should be well-modulated. Though faceted classification scheme is more appropriate for deriving chain, however it is not impossible in case of enumerative scheme of classifications; and

(d) The terms denoting the class number should be general and currently in use.

 

2.4  Method of Chain Procedure

 

First of all the links are derived from the digits of the class number i e first link out of the first digit, the second digit out of the first two digits, the third link out of the first three digits, and so on, upto the last link which is to be made of all digits. Secondly they will be written one below the other in succession and against each link is written its translation into natural language. While translating into natural language, we can use only those terms which are given in the classification schedule. This is done to achieve consistency and uniformity. After putting the natural language, these links are connected with the sign equal (=). Thirdly, this sign of each link is connected with the next succeeding link by a downward arrow. Putting the sign of arrow is merely to make the chain visible to the eye. For example, the class number O111, 1M88, 1 for „The Wasteland‟ by T S Eliot, the following links will be drawn:

 

O = Literature
O1 = Indo-European Literature
O11 = Teutonic Literature
O111 = English Literature
O111, = Connecting symbol of Personality
O111,1 = English Poetry
O111,1M = English Poetry of Poets born in 1800‟s
O111,1M8 = English Poetry of Poets born in 1880‟s
O111,1M88 = English poetry of Poet born in 1888 i e Eliot
O111,1M88, = Connecting symbol of Personality Facet
O111,1M88,1 = The Wasteland by T S Eliot

 

After the links are drawn, the next step is to select the false, unsought and sought links. If the link ends with a connecting symbol or a signature symbol or Time Isolate representing Time Facet of the class, is called as false link. There can be some unsought links in the chain. The decision of a link whether sought or unsought is arbitrary. It depends upon the library which may define it in the light of the reading materials organized, the interest of the readers served and the nature of the service attempted. All remaining links which are neither false nor unsought are known as sought links. These are used for giving subject headings in the class index entries of the alphabetical part of the classified catalogue. In the above example, there are 11 links. But only 1,4,6,9 and 11 are significant or sought links, number 2 and 3 are unsought and 5,7,8 and 10 are false links. The sought links are written from the last link. They are denoted by the terms used in the schedule of classification. They are illustrated below:

 

1 WASTELAND, ELIOT (T S) (1888). O111,1M88,1 O111,1M88
2 ELIOT (T S) (1888), POETRY. O111,1M88
3 POETRY, ENGLISH. O111,1
4 ENGLISH, LITERATURE. O111
5 LITERATURE. O

 

At every progressive step in the chain, the right hand digit is left and the index term is in each case, the term which is denoted by the new last digit.

 

2.5  Class Index Entries

 

After determining the sought links, the next step is to prepare class index entries. In the above case, we have to prepare 5 class index entries derived from the class number O111,1M88,1. One class index entry is illustrated below:

 

WASTELAND, ELIOT (T S) (1888).

For  documents  in  this  Class  and  its  Sub-

divisions, see the Classified Part of the catalogue

under the Class Number  O111,1M88,1

 

On the same pattern the remaining 4 more class index entries will be prepared. There is no difficulty in preparing these index entries after determining the sought headings. In the case of a Commission /Committee

 

Report, two entries are prepared from the last link using as heading each of the following:

 

(a) The words in the heading of the Main Entry (omitting the information about the Chairman of the Commission/Committee), as Main Heading; and the term, such as „Report‟ forming the title, as Subheading; and

(b) The word-group consisting of the entry element in the name of the Chairman followed by the word „Commission‟ or its equivalent and the year used in the

 

Main Entry as the individualizing element, as Main Heading; and the term, such as „Report‟, forming the title, as Subheading.

 

2.6  Advantages of Chain Procedure

 

1. Economical method: Chain procedure is an economical method. This economy is of 2 types:

 

(a) Chain procedure saves the time for compiling and publishing huge subject heading lists, which, even when complete, do not suit the needs of every library, and cataloguers have to mutilate and adopt these according to their needs.

(b) It is a very economical method of arriving at subject headings. Cataloguer‟s time is saved a lot as he need not search the specific subject of the book. Economy in the number of entries may also be secured. According to

 

Coates, “Chain procedure is substantially more economical than other possible systems because it eliminates permutation of components, so that for any particular compound subject, the saving is a question of the ratio between the sum of the components and their factorials. For terms of one or two components there is no saving; for terms of three components (six permutations) there is a 50 per cent saving in number of entries required; for terms of four components (twenty four possible permutations) there is a saving of 83.33 per cent, and for longer compounds the economy is even greater”.

 

2. Mechanical nature: The second claim made on behalf of the chain procedure is that it provides a simple mechanical way. In lists of subject headings, the cataloguer has to use his knowledge and discretion to finalize the number of entries to be prepared for a specific title. But chain procedure is just like „Indian Magic‟. One reached the relevant link and the subject heading is ready. Only you have to use your judgement whether the subject heading is sought or unsought.

 

3. Uniformity and accuracy: The chain procedure is an accurate method. Wherever it is applied, it will give the accurate and uniform subject heading no matter whosoever is the cataloguer, while in the case of complied lists of subject headings, it is some time different.

 

4. Symbiosis between classification and cataloguing: There is a remarkable coordination between classifier and cataloguer in this method. A class number is assigned by the classifier, and the cataloguer turns it into subject headings of the terminology of classification schedule. In case of any mistake on the part of classifier in assigning class number, it is also corrected.

 

5. Alphabetic key: It provides an alphabetical entry under the idea denoted by every digit in the class number. The alphabetical subject headings derived by chain procedure lead the reader to the exact spot in as near to its formulated heading and to his actual need as possible. The subject headings derived on the basis of chain procedure forms therefore a never failing key to the classified part of the catalogue.

 

2.7 Limitations of Chain Procedure

 

Inspite of all its qualities, this system has limitations of its own. They are:

 

1. Cataloguers are deprived of the opportunity of studying books from the point of view of specific subject.

2. If the classifier has allotted the wrong class number, and the cataloguer fails to detect it, the defective subject entries will be prepared.

3. Chain Procedure is entirely dependent on the classification scheme and as a result it can be only as good as the classifications scheme used. For the success of this method the class number assigned to documents must be co-extensive with their subjects and properly modulated. Classification schemes do not always possess this quality.

4. There may not actually be a separate document in the library dealing with the subject represented by one or more of the sought upper links in a chain, which may confuse the readers.

5. The sequence of terms in a heading is not always desired one. The readers mostly consult the concrete idea among the group of words in a compound heading, but in Colon Classification only abstract idea is used, i e

 

BIBLIOGRAPHY, SHAKESPEARE (William) (1564).

 

For  documents  in  this  Class  and  its  Sub-divisions, see the Classified Part of the catalogue under the Class Number O111,2J64a but readers would like to see entry under SHAKESPEARE (William) (1564),BIBLIOGRAPHY.

 

According to E J Coates, its (Chain Procedure‟s) importance lies in the fact that it is the first systematic procedure laid down for subject indexing. It is nearly, though not quite, mechanical in its method of working. It scores heavily over earlier practices on grounds of economy. It provides entry word approach for compound subjects through a combined system of generic and specific entries, but it has the disadvantage that the components in its compound subject entries are not given in the order in which most enquirers will think of them. Finally, it demonstrates subject relationship by alphabetical collection of entries on different aspects of the same subject. It is limited in this respect only by the extent to which the Colon Classification fails to „modulate‟ fully in the listing of terms in hierarchy. The very fact that the British National Bibliography has employed Chain Procedure in preparing its subject index is a scintillating tribute to Chain Procedure‟s superiority over other techniques of subject indexing. Till classified catalogue is in use as a tool for the access of library‟s collection, Ranganathan will be remembered for maximizing the efficiency of this tool by devising chain procedure.

 

3. POPSI (Postulate-Based Permuted Subject Indexing) 3.1 Introduction

 

Chain Procedure devised by Ranganathan has been a significant landmark in subject indexing both in India and abroad. Its major weakness has been its dependence on a scheme of classification. The Class Number is the basis of Chain Procedure. If the Class Number of a document is constructed on the basis of a structurally defective scheme, the Chain Procedure cannot index the document properly. It also suffers from the problem of disappearing chain. The information scientists at DRTC developed an indexing system free from these defects which is called as POPSI. It does not depend on the Class Number but it is based on Ranganathan‟s postulates and principles of general theory of classification. Hence it is considered an improved version of Chain Indexing.

 

3.2  Advantages of POPSI

 

POPSI is helpful in:

 

  1. Formulating subject headings which may be used as feature headings or for other indexing purposes.
  2. Deriving subject index entries for a classified index, or for an index to a book, etc.
  3. Determining the subject of a reader‟s query in a consistent and helpful way.
  4. Formulating a strategy for searching information about a subject in a catalogue or other surrogate files, and
  5. Deriving a base for the presentation of ideas in the text of a document.

 

Though POPSI is basically designed for manual methods, it is amenable to computers for which an improved version of software package is being developed.

 

3.3  Steps in POPSI

 

1. Verbal representation

2.Display of components

3. Short display

4. Approach terms

5. Deriving subject index entries

6. Display of subject index entries

7. Cross references

8. Alphabetical arrangement of entries

 

3.4  New Version of POPSI

 

This version of POPSI recommends a set of elementary categories which

are:

 

Discipline (D), Entity (E), Action (A) and Property (P)

Discipline (D) covers conventional field of study e.g. Physics, Medicine, Political Science etc.

Entity (E) includes manifestation having perceptual correlates-Energy, Plants, Place, Time etc are its examples.

Action (A) includes manifestations denoting the concepts of „doing‟.

Property (P) includes manifestations denoting the concepts of „attribute‟, e.g. Effect, Efficiency, Power, Property etc.

 

3.5  Working of POPSI

 

The working of POPSI can be understood with the help of an example:

Document: Treatment of Tuberculosis of Lungs

It may be analyzed as:

First step:     Verbal representation

D = Medicine

E = Lungs

P of E = Tuberculosis

A = Treatment

Second step:  Formalization of the sequence of components:Medicine (D), Lungs (E), Tuberculosis (P of E), Treatment (A on P).

Third step:   Standardization and is concerned with semantics:Medicine (D), Lungs (E), Tuberculosis (P of E), Treatment (A on P).

Fourth step: Modulation with augments:Medicine (D), Man, Respiratory system, Lungs (E), Disease, Tuberculosis (P of E), Treatment (A on P).

Fifth step:     Preparation of the Entry for organizing classification:Medicine 6 Man, Respiratory system, Lungs 6.2 Disease, Tuberculosis 6.2.1 Treatment.

Sixth step:   To decide the terms of approach

Treatment See Therapeutic

 

In the above chain, general terms like „Medicine‟ and „Man‟ are not selected as approach terms.

 

Seventh step: To prepare entries under each term of approach as given below:

Therapeutic

Medicine 6 Man, Respiratory system, Lungs 6.2 Disease, Tuberculosis 6.2.1, Therapeutic.

Tuberculosis

Medicine 6 Man, Respiratory system, Lungs 6.2 Disease, Tuberculosis 6.2.1, Therapeutic.

Similarly there will be entries under Disease, Lungs, Respiratory system.„Man‟ and Medicine will be left out.

Eighth step: Alphabetisation of all the index entries and references.

 

3.6  Potentialities of POPSI

 

It can be used in various ways (i) to construct subject headings (ii) to generate subject index entries for classified catalogue etc. It is being used in a number of bibliographical and indexing publications.

 

Conclusion

 

POPSI is an extension of Chain Indexing. It has solved the problem of disappearing chain which was major criticism against chain indexing. It made the indexing system free from classification scheme. Though it is designed for manual methods, it is amenable to computer use also.

 

4. PRECIS (Preserved Context Indexing System) 4.1 Introduction

 

PRECIS came into existence in 1968. It has been developed by Derek Austin. It was the result of long research which was undertaken by the Classification Research Group (CRG, London). After testing it thoroughly, it was introduced in BNB in 1971. The BNB since its inception in 1950 has been using Ranganathan‟s Chain Procedure for preparing its alphabetical index. It is supposed to be free from all the limitations which chain procedure was suffering from. It may not be the best indexing system but it is believed to be better one.

 

4.2  Definition

 

The originator of PRECIS, Derek Austin has defined it as a system, “in which the initial string of terms, organized according to a scheme of role operators, is computer-manipulated so that selected words function in turn as the approach term. Entries are restructural at every step in such a way that the user can determine from the layout of the entry which terms set the appropriate term into its context and which terms are context dependent on the approach terms”.

 

4.3    Chief Characteristics of PRECIS

  1. It is independent of classification scheme.
  2. The headings to be generated by PRECIS are to be coextensive with the contents of the document at all access points.
  3. The entries generated by the system are to the meaningful so as to enable the users to interpret them correctly.
  4. The system is quite flexible to accommodate the new terms put into the literature freely.
  5. The system is amenable to the use of computer for generating additional index entries.

   4.4 String Making

 

Two principles are followed in the preparation of a string- (i) context dependence, and (ii) one-to-one relationship. This may be understood by taking an example:

 

Recruitment of Staff in the Colleges of Rajasthan

 

Here the component terms are: Recruitment, Staff, Colleges, Rajasthan. Hence we can say that there is a state Rajasthan which has staff, whose Recruitment is the theme of the document. The arrangement of the components in the string will be as follows:

 

Rajasthan. Colleges. Staff. Recruitment.

Here each term is directly related to the next term in the string.

 

4.5 Role Operators

 

The principles of „context dependency‟ and „one-to-one relationship‟ are adopted for the order of terms in the string.

 

Format

 

In the example given above the situation will be as follows:

 

Rajasthan. Colleges. Staff. Recruitment

Or

Recruitment. Staff. Colleges. Rajasthan.

 

The context in the above entries is intact. But this is not so in the following entries:

 

Colleges. Rajasthan. Staff. Recruitment

Or

Staff. Rajasthan. Colleges. Recruitment

 

4.6  Generation of Entries

 

Index entries are generated from the string by „shunting‟. Each component term in the string becomes an approach term for the users by turn. This portion is shown below:

 

Rajasthan. Colleges. Staff. Recruitment

Rajasthan

Colleges. Staff. Recruitment

Colleges. Rajasthan

Staff. Recruitment

Staff. Colleges. Rajasthan

 

Recruitment

Recruitment. Staff. Colleges. Rajasthan

 

4.7  PRECIS V/s Chain Indexing

 

String formation in PRECIS is not dependent on classification scheme. Hence it is not affected by the structural defects of the classification scheme. It has also got over the problem of the disappearing chain which is a major point of criticism against Chain Indexing. It does not face also the problem of empty links as is the case with Chain Indexing.

 

Chain procedure is all right for a classified index but not so effective for alphabetical index. To make it suitable for alphabetical index certain modifications are suggested. POPSI is considered as the modified version of Chain Procedure which has been developed by DRTC. In it the problem of disappearing chain has also been solved.

 

Summary

 

PRECIS has also been adopted by national bibliographies other than BNB and indexing systems. In the words of Phyllis Richmond:

 

“The PRECIS system is one of the most versatile. In general, one may consider it a quantum jump ahead of contemporary competition”.

 

PRECIS has been criticized as it has not been able of have both the semantic and syntactic roles of terms in one single string. But a final judgement about its proficiency can not be passed.

you can view video on Subject Cataloguing: Chain Procedure, POPSI and Precis

Further Readings

  1. Austin, Derek: PRECIS. Lib. Sc. with a slant to Doc. 12(4), 1975. 89-126.
  2. Job, M.M.: Chain procedure. Herald of Lib. Sc. 12(2-3), 1973. 188-193.
  3. Kaula, P.N.: Chain indexing. Herald of Lib. Sc. 9(4). Oct, 1970. 318-325.
  4. Prasher, R.G.: Index and indexing system. Ludhiana, Medallian Press.
  5. Thirumalaimthuswamy, A.: Chain Precedure. Lib. Herald. 7(1). April 1964. 71-79.