Introduction

Publishing has traditionally been regarded as a profession for gentlemen. This is an outmoded term to be sure, but it conveys the impression of a trade that was not overly concerned about profit, of practitioners focused on some higher ideal. Librarians have also traditionally exhibited this trait of not focusing on money, perhaps by emulation, perhaps by an innate commitment to collaboration and the ideal of learning. This is epitomized in the title of the library journal devoted to the commercial nexus linking libraries, publishers and intermediaries, Against the Grain, which implies that its subject matter is somehow at odds with traditional professional concerns.

In contrast to the USA, such distaste may have been fostered by the long existence of retail price maintenance for books in many European countries, such as Austria, France, Germany, Greece, Holland, Italy, Portugal and Spain. In Germany, for instance, publishers are legally required by the Buchpreisbindungsgesetz to set the price, including tax, at which books are sold to the end-user. The price remains fixed for a minimum of 18 months, or until a new edition is published.

The UK form of retail price maintenance, the Net Book Agreement (NBA), enabled publishers to fix the prices at which books could be sold. Booksellers were not allowed to give discounts, except a fixed 10% discount under the so-called Library Licence. Libraries received the fixed discount on condition that they made their stock available to the public. This condition, obviously fulfilled by public libraries, was generally not considered too onerous by university librarians: virtually all allowed walk-in access to stock. The notable exception was the poet Philip Larkin, who as Librarian at the University of Hull from 1955–85, would not allow students of the local teacher-training institution access to the university library, and hence forfeited the 10% discount (Hinton, 1987, 45), an example of the triumph of principle over economics.

The early 1990s saw the discounting of book prices by some of the larger bookselling chains, by the mid-1990s publishers were withdrawing support and in 1997 the NBA was ruled illegal.

Introduction

As illustrated in the other chapters of this book, readers’ expectations and the collection and development strategies of libraries are changing in this digital age. The move to a greater reliance on electronic information is influencing how society searches for and uses information and is impacting on how libraries and their collections are viewed and used.

In the public arena, the British Government is keen to provide more and more information on the internet. This ranges from local government to health information, and how individuals can get involved in the ‘Big Society’. To this end, the British Government launched two initiatives. The first was the Information Matters strategy, which recognized that ‘More information is being created and held today than ever before. It is the lifeblood of virtually every service we use as citizens, whether run by the public or private sector’ (Knowledge Council 2008, 2). The Information Matters strategy aims to enhance and create standards on how government departments find, use, store and make information available. The strategy promotes the concept that government staff had responsibilities as ‘creators, custodians and users of knowledge and information’ (Knowledge Council, 2008, 3) and that they required skills and tools to meet these responsibilities. The second was the ‘Digital Britain’ initiative. The Digital Britain Interim Report highlighted the need for Britain to develop both the technological and skills infrastructure to succeed in the digital economy. The report argued that for the UK to ‘maximise the digital opportunity, we will need to ensure a population that is confident and empowered to access, use and create digital media’ (DCMS & DBERR, 2009, 66). The report of the Digital Britain Media Literacy Working Group (Ofcom, 2009, 9) divided the skills individuals need to operate in today's society as digital life skills and digital media literacy skills.

In academia, there has been a radical shift from print to electronic information in terms of electronic searching tools as well as the electronic full text of journals and textbooks.

The most superficial examination of a library, bookshop or other collections of documents will reveal that the idea of place is a very common element in the subjects of books. Titles abound such as:

It is therefore vital that any system of subject cataloguing enables us to express these geographic aspects. LCSH does this through the medium of geographic subdivisions.

When to use geographic subdivisions

The great majority of headings are accompanied by the legend (May Subd Geog), standing for ‘May Subdivide Geographically’, which tells the cataloguer that the name of a place can be added to the heading.

Where there is no such instruction, you should assume that geographic sub-division is not permitted, and sometimes, in case you were in doubt, there is a specific injunction not to subdivide.

Example

This is usually because such subdivision would be unnecessary, the topic being already limited geographically, as with names of places, or because alternative headings make provision for local specification, as in Prisoners’ writings, Bulgarian or Prisoners’ writings, Chinese. That is not always the case, however, and in some cases it is very hard to see why geographic subdivision is forbidden. Some general types of headings seem always to be debarred from geographic subdivision: for example, names of families, names of fictitious and legendary characters, and headings of the type Brooms and brushes in art, Bureaucracy in motion pictures or other similar combinations.

The preceding chapters have been mainly concerned with how the subject headings are presented, and what form they take. The form of headings is decided by the Policy and Standards Division of the Library of Congress, and you as cataloguer don't have any control over this; individual headings must be used consistently by everyone, both at the local level to ensure efficient retrieval, and across institutions to support the exchange of information and interoperability of catalogues.

In selecting headings for individual documents, however, you have much more freedom to do as you think best for your own situation. In this and the following chapter we shall look at how to choose headings in order to represent documents accurately and in a manner useful to end-users.

Before you can do that it is necessary to decide what the item being cata - logued is about. Whatever system of subject headings (or classification scheme or thesaurus) is being used to describe a document, you should try initially to make an independent assessment of what the subject of that document is. In practice, you will almost certainly be unable to represent this exactly using the artificial language of your system, but you should at least begin by deciding objectively what it is you want to express. This process may be called ‘subject analysis’, or ‘document analysis’, ‘content analysis’ or ‘concept analysis’. The subject content of items is sometimes also referred to more grandly as ‘intellectual content’ or ‘semantic content’, but these are simply other ways of defining what a document is about.

The problem of ‘aboutness’; indexer consistency and Subjectivity

The process of determining ‘aboutness’ is an essential part of subject cataloguing, but it contrasts markedly with the task of descriptive cataloguing. In the majority of cases the cataloguer has only to inspect the title page to determine the author and the title, and other bibliographic details of a book, and having found the author and title there is generally no dispute as to what they are. In some cases the subject is similarly straightforward, but in others the situation is more complicated. Unfortunately there is nowhere to look in a book which will tell you plainly and unequivocally what its subject is; you must decide this for yourself.

The electronic LCSH is the most significant digital tool for the cataloguer, but there are other online versions of LCSH, and ways in which Library of Congress support the work of subject cataloguing generally in a free-to-access manner. The most important of these is the catalogue itself.

The Library of Congress catalogue

Various references have been made to the catalogue of the Library of Congress throughout the text, but it seems sensible to bring them together here. The very existence of the catalogue is a major contributor to the increase in the usage of LCSH, through the easy availability of bibliographic records and their subject cataloguing data. The catalogue is a substantial resource for subject cataloguers in three major ways: the records themselves, the subject index and the authority records.

Accessing the LC catalogue: bibliographic records

A link to the catalogue is to be found on the Library's home page at www.loc.gov. (For UK users it may be worth noting that access to the catalogue is very quick and easy during the morning, but it becomes busier during the afternoon, and you may need several attempts to log in then. Weekends are not usually busy, and although there are occasional periods when the catalogue is down for routine maintenance, these don't usually happen at inconvenient times.)

The home page also provides links to other useful pages, including a number of services and resources specifically for librarians, such as information about MARC, thesauri and vocabulary, metadata standards, and the Cataloging Distribution Service, which markets LC products. Following the link to the catalogue gives you a choice between quick search, basic search, and guided search (Figure 18.1).

Basic search (Figure 18.2) offers a good range of search options and, apart from the most specialized tasks, is the most useful approach for finding records. There are some tips for searching at the bottom of the screen.

Search options include keyword search in all the major aspects of the record (author, title, subject and series). You are limited to a single keyword for each of these, and searching for combinations of keywords requires the use of the Guided Search facility.

Free-floating subdivisions are non-geographic concepts that can be added to headings. Until 2003, they were contained in the green Subject Headings Manual, but they are now included in the red books; they are also available in a separate print publication, Free-floating Subdivisions: an alphabetical index, and they are searchable as a separate database within the online LCSH. The print version is now in its 23rd edition (2011), and unfortunately this will be the last.

We saw in Chapter 6 on content analysis that some varieties of concept occur very frequently in the subjects of documents: concepts such as place, time, form and persons, the last often as the intended readership. In classification systems these very common ideas are usually contained in special auxiliary tables, represented by codes which can be attached to classmarks, sometimes using particular symbols or facet indicators to indicate that an auxiliary concept is present. In LCSH a similar purpose is served by the free-floating subdivisions. The idea of the free-floating subdivision is that it can be used without any specific instruction, and attached to main headings as required. Some can be added to any heading, whereas others are restricted to particular types of heading, such as those for individual authors which are used to specify the different texts and editions and secondary publications associated with an author.

Free-floating subdivisions are added to main headings using the double dash in exactly the same way as topical and geographic subdivisions. Conveniently, the dashes are included in the subdivisions as they are displayed in the alphabetical list in Classification Web. The subdivisions are also accompanied by copious notes explaining the way in which individual subdivisions are to be used:

–Classification

–Editions, Curious

–Kings and rulers

–Semantics

Apart from the structured headings constructed from the elements of the main headings and various kinds of subdivisions, there is only one kind of heading that can be created by the cataloguer: the name heading.

What constitutes a name heading?

Name headings are used when some person, place or entity with a proper name (a name beginning with a capital letter) occurs as the subject of a work. You need to make a careful distinction between the person or entity as the subject of the work, and the other ways in which they might be represented in different parts of the catalogue or metadata record; for example, a person may be author, editor or illustrator, and an organization could be a corporate author or a publisher. Only when the document is about the person or organization should a subject heading be made.

Because there are so many possibilities for name headings, LCSH does not routinely include these in the published list of headings, but expects the cataloguer to create them as needed, following certain rules and conventions for their construction. Notable exceptions include prominent historical figures and prolific authors and artists, the latter often serving as pattern headings for others in the same category. The Library of Congress Authorities database (http://authorities.loc.gov/) provides a useful check for practitioners as to the forms of names that have been used there, and can be searched freely online.

There are several different categories of things that can have name headings. I have taken a fairly liberal view of what constitutes a name heading, and Library of Congress would regard some of these technically as title headings (which are used for created works such as works of art or literature). My general policy has been to embrace all those unique persons or objects for which headings can be developed by the cataloguer where they do not exist in the published list; it seems important to address the situation where a heading for such an entity isn't available, and the cataloguer wonders what to do about it. Everything in this chapter, therefore, is indicative of the kinds of subjects where cataloguers can adopt an innovative stance and exercise their own judgement in ‘making up’ a heading, using comparable examples as a model.

Because LCSH is such a complicated system, we will start by looking at how you can apply the headings exactly as they are published in the lists, either printed or online. In lots of cases the subjects of documents can be very well expressed simply by using one or more headings, without any further effort. In later chapters we will consider how to add subdivisions to these main headings, and how to create new headings where this is permitted. If you have access to Cataloguers’ Desktop, it would useful to read the Introduction to the Subject Headings Manual at this stage. [Note: I have used the expression ‘main headings’ to indicate the headings as they occur in the published list of headings, and to differentiate these from subdivisions, and combinations of headings and subdivisions.]

Content analysis and LCSH

It is important that you go through the process of analysing the content of the document and establishing its subject before you begin to decide which headings to use. Remember that content analysis has two main objectives: to make a concise summary of the document's subject; and to identify any topics that users may be trying to retrieve. In the first case, the concept analysis should state the general subject of the document, which may be very broad (economic theory, the postal service, archaeology), or quite specific and complicated in nature (American films of the 1930s, manuscript illumination in Medieval Germany, laser excision of epithelial tumours). Your objective then is to try and find a heading, or set of headings, that will represent all the significant parts of the document's subject.

Analytical cataloguing

Most conventional library cataloguing and indexing stops at that point. The document is located physically alongside other documents with similar and related topics, and it has sufficient data in the subject field of the catalogue record to retrieve it according to its content. In other environments the practice of analytical indexing or cataloguing aims to provide subject information about the detailed content of a document; individual chapters or sections are indexed, and there is much more information about very precise topics.

Early subject headings

In the nineteenth century such library classification schemes as existed were very rudimentary, often consisting only of a few broad subject divisions with the books then arranged by size or author's name, or some other principle unrelated to the subject. Locating a book on a specific topic could be a very hit and miss affair, as the lack of detail in the classification meant that related material was usually scattered throughout the relevant class. The more usual method of subject searching was through alphabetically arranged subject catalogues. These might take the form of large ‘guard books’ or ledgers into which the catalogue entries were written or pasted, and which were commonly reproduced as printed books. The alternative form of card catalogue came into common use in British and American libraries from the 1870s onwards.

The earliest subject entries were usually simply names of topics, such as ‘Geology’ or ‘Cricket’. In many cases the subject keywords or ‘catchwords’ were taken directly from the title of the book or article. Where necessary the subject headings might be combinations of two or more words, for example, ‘Military history’ or ‘Education of women’, and in some early subject catalogues and indexes headings can be found with subdivisions, rather in the manner of book indexes.

Whatever the format, most of these early headings were created on an impromptu basis, and there were no general rules applied to ensure that the format of headings was logical or consistent. It often appears that not much common sense was exercised in their selection, nor thought for what a reader might reasonably look up.

Cutter's Rules

The idea of systematic subject headings is usually credited to Charles Ammi Cutter, an American librarian, who, in 1876, the same year that Dewey created his Decimal Classification, published a slim volume called Rules for a Dictionary Catalog. This was one of the earliest comprehensive statements of theoretical principles for making entries in a catalogue, although several libraries had produced some sets of cataloguing rules during the mid-nineteenth century. Most of Cutter's Rules were concerned with how to enter the details of the author and title of a work, and what are known as the descriptive elements (the physical dimensions of a book and the number of pages, for example), but he also provided some guidelines on how to make subject entries in a catalogue.

Today searching for information using words and phrases from natural language has become the norm. Everybody knows how to use a search engine, and most people will start a search by entering the name of whatever it is they wish to find out about. Although directory style methods of organizing information on websites are also common, many searchers find it simplest and easiest just to put some appropriate words into a search box. The existence of so much information in electronic format reinforces this approach, since it is easy to process huge quantities of text very rapidly, mechanically matching strings of characters to locate and retrieve the sought terms.

This increasing preference for language-based retrieval is mirrored in managed information resources, databases, indexes and bibliographies, and library catalogues. While the classification scheme was the dominant means of organization and retrieval in libraries for two-thirds of the twentieth century, with the arrival of the first automated systems in the 1950s and 1960s, there was an accompanying move towards language as the basis of indexing, and the emergence of tools such as keyword lists and thesauri. These have continued to be the norm for electronic collections or for collections where physical access and browsing are not important. In the library context, where automated retrieval by means of an online catalogue is combined with a substantial physical collection of documents, the subject heading list has usually been considered a more appropriate system.

This is particularly noticeable in the UK, where the use of subject headings in library catalogues has become widespread over the last 20 years. A number of factors have doubtless contributed to this situation: greater awareness of, and access to, subject headings through the medium of web-based catalogues; the advantages of bibliographic information sharing made possible by this phenomenon; the consequent need for standardization, particularly of the bibliographic tools used; and the ease of retrieval using words, as noted above. Why the subject heading list, rather than the thesaurus, should be preferred is not altogether clear, but it is now very well entrenched in most academic libraries of the developed world, the UK being one of the few countries not to have its own national system of subject headings.

Classification Web is the Library of Congress's electronic version of the classification and subject headings. It is also sometimes referred to as Classweb.

Classification Web is only available by subscription, the price of which is related to the number of simultaneous users. This does tend to make it less accessible to small libraries, and unfortunately there is no educational rate. Nevertheless, for libraries where the budget can accommodate it, it offers a quick and easy way to use LCSH with all the advantages that a digital format affords. As the system is updated daily, it also provides the most current version.

The composition of Classification Web

Classification Web gives you access to both the Library of Congress Classification and the Library of Congress Subject Headings. We won't consider the classification specifically, but there are several ways in which the two are linked together that will be dealt with in this chapter.

As far as LCSH is concerned, the online tool consists of several distinct databases. The equivalent of the printed LCSH (the red books) is held in the main database, and this is what you will normally search or browse. Headings for particular forms and genres can be searched as a separate collection, although form and genre as subject are represented in the main list as well (the distinctions here are treated in more detail below). There is also a separate database of the free-floating subdivisions, which can be searched independently, or in combination using the advanced search. You can also look for name headings, which are derived from the Library of Congress name authorities. That is a useful device for establishing the correct form of personal and corporate names without having to search the name authorities as a separate exercise.

The classification data for the Library of Congress classification also forms part of the system, and various elements of LCSH link to that data, and to Library of Congress catalogue data. We will see below how this all works together.

Layout and format

Although the content of Classification Web's LCSH is to all intents and purposes the same as the printed version, it looks rather different on the screen. There are also some differences in the way structured headings are presented.

In disciplines such as literature, music, and the visual arts, the material to be dealt with can be divided into two major groups: the works themselves, poems, novels, scores, recordings, and so on (primary sources); and works of interpretation and criticism (secondary sources).

In the arts, organization by subject content is seldom seen as the most useful means of arranging resources; arrangement by language, form, period, or genre (or rather, combinations of these) provides an easily understandable and non-contentious method, and is preferred at least for initial organization. It also reflects the way in which literature, for example, is commonly studied.

Although LCSH is not used to provide a systematic arrangement, this focus on attributes of language, form, period and genre is reflected in many of the headings for the creative arts:

A great number of the headings for literature reflect this basic preoccupation with the three major facets of language, period and form, but LCSH has a considerable advantage for retrieval over its associated classification scheme (and indeed most other knowledge organization schemes in their treatment of literature). LCSH allows the cataloguer to express an unusually large range of attributes of literature, including very precise themes and topics, as well as literature specified by all sorts of social and cultural characteristics. It can be very erratic in coverage and, as usual, very unpredictable in what is included and how it will be manifested, but you should find yourself able to express some very complex ideas through the different kinds of headings.

Headings for literature: language, period, and form

The examples at the beginning of this chapter demonstrate a degree of precoordination, which is a general characteristic of the headings for creative subjects, and this is certainly the case for literature. There is fairly comprehensive coverage of languages and forms in combination (e.g. Swedish fiction, Urdu drama) using headings of this general form.

In the previous chapter we looked at how to select individual headings for an item. Nearly all of the headings were pre-coordinated combinations of two or more concepts such as Chocolate factories or Equine dentistry, and indeed it can be hard to find headings that represent only simple concepts. Even so, in the case of more complex subjects, a series of headings is usually necessary to cover all the various aspects of the subject. In many cases complicated subjects can be expressed more elegantly by structured or composite headings rather than by a list of individual headings.

The nature of structured headings

Although cataloguers cannot invent headings, they can build extended headings using subdivisions provided as part of LCSH. (In Classification Web the subdivisions are held in a separate database from the main headings, and can be searched for independently.) The subdivisions are added to the main headings, subject to certain rules and constraints. These extended headings may be referred to as structured headings since they consist of more than one component or element. Like many other features of LCSH, structured headings can only be created according to the LCSH rules, and you should not be tempted to invent new ones by analogy.

For the most part, structured headings of this kind are not routinely included in the published LCSH, although you will come across some examples. This means that there are very many potential headings that individual cataloguers can create as needed.

Sometimes structured headings can become very complicated, with five or six elements, although a combination of two or three is more common. Typical examples of structured headings are:

Bells–Collectors and collecting

Cosmetics—Additives

Morse code–Problems, exercises, etc.

Organometallic chemistry–Nomenclature–Handbooks, manuals, etc.

Parapsychology and crime–Case studies

Perpetual calendars–Belarus—Periodicals

Books and reading–England–Northumbria (Region)–History –To 1500—Sources

It feels somehow more satisfying to express a compound subject with a compound heading, and a single heading of this type seems to me essentially better than a list of unconnected headings, since it provides a succinct statement of the content of a document.