A few Blair babes but not a lot of mickle

We all know that scholarly plagiarism is endemic in lexicography, but we may not have realised that it can be raised almost to the level of an academic genre in its own right. The demonstration is provided by this offering from Oxford University Press, which mimics one of the most original dictionaries of the 19th century, even down to cribbing the title.

Ebenezer Cobham Brewer, who conceived the idea of a dictionary of "phrase and fable" and published his own remarkable one-man compilation in 1870, was the son of a Victorian schoolmaster. In a generation thirsting for education, he realised that many members of the rising reading class were often hampered by their very limited - sometimes non-existent - acquaintance with earlier literature, and consequently found themselves stumped when they encountered in English writers' references to the heroes and mythology of antiquity, or the learned use of expressions from Latin, French and other languages. His Dictionary of Phrase and Fable was an instant success in an age when affordable works of reference were few and far between. Brewer's compendium sold 100,000 copies, and he brought out an enlarged edition in 1894. Reprints and revisions have continued to appear ever since.

Elizabeth Knowles, who has edited other dictionaries for the OUP, acknowledges a debt to the "original conception" of Brewer's publication, while pointing out that today any work following in Brewer's footsteps faces "different challenges". But she has little to say about what these challenges are or how she has tried to face them.

The resuscitation of Brewer's now archaic-sounding title itself poses problems. Neither Brewer nor Knowles offers us a definition of what a "phrase" is. The majority of entries in both volumes turn out to be single words, often names.

When Knowles does list a phrase, she often seems, unlike Brewer, more interested in dating it than explaining what it means. All she tells us about "many a mickle makes a muckle" is that it is a proverbial saying from the late 18th century, whereas "many a mickle makes a little" dates from as long ago as the 13th. Baffled, we look up the word "mickle", only to find ourselves referred back to the two phrase entries listed above. As a last resort we look up "muckle"; but for "muckle" there is no entry at all. Now it strains credulity to suppose that the average reader who finds it necessary to look up all this business about mickles and muckles is quite clear about the meaning, but merely desires to know how many centuries ago such expressions were coined.

By comparison with Knowles, Brewer is a beacon of enlightenment. He not only tells us what "many a mickle makes a muckle" means, and for good measure why the version "many a little makes a mickle" makes no sense at all, but also cites comparable proverbs from French and Greek. Here as elsewhere, Knowles gives the impression that she has failed to take the needs of foreign students of English sufficiently into account, let alone those natives who may be less than familiar with all the dialectal varieties of the British Isles.

As regards "fable", the situation is even more curious. Although Brewer never explains what a fable is, he at least provides a list of "the most famous writers of fables" in many lands. Knowles gives a rather perfunctory definition of "fable". But if you want to know anything about who wrote them, you will have a hard time digging it out. She does give Aesop a brief entry, but La Fontaine and Lessing are nowhere to be seen.

Brewer is at his best in an article such as "dog", which he divides meticulously into 11 parts. These concern dogs of note, dogs of noted persons, dogs who were models of their species, dogs in phrases, dogs in metaphors, dogs in the Scriptures, dogs in art and so on. Nothing as systematic appears in Knowles. She apes Brewer by providing a list of famous dogs and their owners (real or fictional), but then disperses the rest of the doggy domain across umpteen separate entries and cross-references. This fragmentation, apart from being infuriating for the user who wants to check them all out, misses drawing our attention to the subtler cultural implications of the concept "dog" that Brewer's more methodical approach manages to capture.

Knowles's etymologies are better than Brewer's, as indeed they should be after another century of philological research and the archives of the Oxford English Dictionary to draw upon. But when she includes miscellaneous items of historical, biographical and scientific information on the ground that they are "central to the development of a civilisation or culture" the effect is to turn her dictionary into a poor man's encyclopedia. At this point a different set of criteria necessarily comes into operation, and these criteria have little to do with either phrase or fable. It is difficult to have much confidence in an encyclopedia that implies that we shall be missing something central about the development of western civilisation if we are deprived of an entry on "Blair's babes", but that, on the other hand, can happily give us potted biographies of Franklin D. Roosevelt and Joseph Stalin while conspicuously omitting Winston Churchill.

If we can forgive Brewer his omissions, biases and inaccuracies, it is because we recognise in the rough edges of his work the condensation of one individual's lifetime of reading. That gives it what any latter-day computer-aided simulation will inevitably lack: the stamp of personality.

Roy Harris is emeritus professor of general linguistics, University of Oxford, and editor, Language & Communication .