Roget’s Thesaurus of English Words and Phrases was first published 150 years ago, the retirement project of Dr Peter Mark Roget (1779–1869), a polymath whose accomplishments included the invention of the slide rule, who wrote a seminal paper that led to the invention of cinema, and whose interests and expertise embraced, inter alia, philology, geology, medicine, laughing gas, Dante and chess. It has proved to be a continuously successful reference work, and an eighth edition, edited by George Davidson, has now been published to celebrate this anniversary.
With its Classes, Sections, Heads, Paragraphs and Groups, the Thesaurus deliberately mimics the Kingdoms, Orders, Phyla, Genera and Species of Linnaean taxonomy in the natural sciences. A national language is not a taxonomy, of course; or rather it is not conceived as one. It does make discriminations and distinctions, though it can tolerate synonymy, since its purposes are many, and ornamental beauty is among them. Words arise from generations of use and misuse; and innovations and inventions in cultural or technical spheres and discoveries in the material world insist on neologism, as do fashion, marketing and journalism (themselves hardly discrete categories). A language, far from being a tool to effect one type of communication, inquiry, or activity, is a record of all such inquiries: it is the product of a series of decisions, shocks, appropriations, assertions, mistakes, beliefs, taxonomies, discoveries, values, ideas, desires.
The old myth that Eskimos had “more words for snow” than the English, though demonstrably false (and careless in its own linguistic failure to discriminate between various “Eskimo” languages), was perhaps successful as a meme because we feel it ought to be true; that a genus of importance will be finely differentiated into species. To this extent, “Eskimo snow” could be a synonym for “what matters to us”.
Language is also less susceptible than the plant or animal kingdom to taxonomic organization; synonymy arises by design rather than classificatory error; homonyms flourish, requiring a single entity to be placed in several different classes; and similarities can be established between elements within otherwise discrete classes and subheads. That said, by bolstering its remarkable systematization with cross-references and an index, Roget’s Thesaurus has proved a serviceable and durable classification, and, for all its artifice and oddity, it is hard to imagine a better one.
To move directly through a headed section, resisting the temptation to take one of the numbered exits, is to see a swatch of linguistic colours flipped quickly before one’s eyes: not all of them will suit your own living room. The words are rarely perfect synonyms but more often are semantic neighbours, differing by minute gradations of sense or tone or implied context; and in any area they give an idea of what, throughout the language’s evolution, an accidental consensus has thrown up as valuable in terms of cultural attention and inattention. They also suggest the restrictions on what we can say: for even if language did not itself shape our thought, then, as any writer knows, one can sometimes find language inadequate or intransigent when giving thought a linguistic form.
In certain disciplines (such as botanical taxonomy itself) specialist vocabularies evolve. Roget, already filled with 300,000 words (by its editors’ estimate), does not contain these words: it restricts itself, mostly, to words in “general” use, though even the word “general” is elastic (and simply looking at one of its entries in Roget demonstrates a semantic range, from “universality” through “broad” and “usual” and “loose” and “vague” to “common” and “vulgar”). Within that “common” frame of reference, however determined, there are a number of terms by which one might make nuanced discriminations: though to call that number “vast” or “generous” or “sufficient” would be a judgment call of its own.
Roget’s ambition was to create a tool whereby accurate expression, rather than ornate and ostentatious speech, was assisted. More loftily, he wondered whether such a book, alongside others in other tongues in a “Polyglot Lexicon”, could move, via a “Philosophical Language”, to a “Universal Language”:
“Nothing, indeed, would conduce more directly to bring about a golden age of union and harmony among the several nations and races of mankind than the removal of that barrier to the interchange of thought and mutual good understanding between man and man, which is now interposed by the diversity of their respective languages.”
It was Roget’s son, John Lewis Roget, who developed effectively the system of cross-referencing that makes the book more responsive to the shifts of language between their original categories. One would be hard pushed to cram every word in the English language into the perfect order of Roget’s primary rationalization; as his son remarked, “the fabric of our language has become a texture woven into one by the interlacing of separate branches, springing from separate stems”. Since 1962, Roget’s original system—to have, as far as possible, a head in one column and its antithesis in an opposing column—dilemmas, one might call them - has also been abandoned, in favour of a continuous listing of headwords: and it is hard to see how the earlier system could ever have worked entirely. The simple example of “convexity” being the antithesis of both “concavity” and “flatness” can illustrate some of the problems here; or the positioning of “indifference” with relation to “desire” or “aversion”.
These examples are Roget’s; he knew full well the difficulties facing his project, and the failures of his attempt, writing:
“In a work of this nature, where perfection is placed at so great a distance … I have thought it best to limit my ambition to that moderate share of merit which it may claim in its current form; trusting to the indulgence of those for whose benefit it is intended, and to the candour of critics who, while they find it easy to detect faults, can at the same time duly appreciate difficulties.”
To read again Roget’s introduction to a work which has changed remarkably little since its first edition is to note, on the one hand, the modesty with which Roget proffered the fruits of several years’ work, and, on the other, the unrealized ambitions he nurtured for it. Its commercial success is undoubted, and its continued vitality assured. But the Universal Language that would have brought harmony is as utopian a dream as ever, and Roget’s assumptions of linguistic competence in the Thesaurus’s audience have been terribly misplaced: despite the expansion of education in England, the functional subliteracy of undergraduates, schoolteachers, broadcasters, professionals and Deputy Prime Ministers gives few grounds for optimism that they can reliably make “the proper choice” with “instinctive tact”, as Dr Roget hoped.
In terms of the limitations of such a work, Roget is frank about the problems of the economy of his classification (for instance, the repetition of terms in more than one heading; the difficulty posed by homonyms; the varieties of neologism). He is right also to stress how rare true synonymy is in English: “it is hardly possible to find two words having in all respect the same meaning, and being, therefore, interchangeable; that is, admitting of being employed indiscriminately, the one or the other, in all their applications.” Readers turning to the thesaurus for a perfect synonym, by which they might add glitter or variety to their language, are, without the help of a dictionary, likely to misrepresent themselves. Where synonymy does exist on a grand scale is in idiom: a glance at Viz magazine’s parody of the Thesaurus (titled Roger’s Profanisaurus) is evidence of a vastly inventive demotic approach to synonyms for body parts and bodily functions. Some are freshly minted, only a few of which will then gain currency; some are specifically regional in origin and cheerfully submitted and thus disseminated by readers. They are colourful, metaphorical, abundant beyond utility, flagrantly uneconomical, wavering between euphemism and dysphemism and, as such, evidence of a combination of relish and disgust.
Profanity is a pool of language that the Thesaurus treads carefully around and only shyly dips a toe into. It is not that the vulgar is wholly excluded: Roget himself spoke up for it, and for idiom, but for him “slang” was the closest to “vulgarism” that he cared to go: profanity and obscenity seem to fall outside. Wavering, like any lexicographer, between description and prescription, his sense of “taste” and “aptness” forever vulnerable to the decisions of the general population, he noted: “I did not feel myself justified in excluding them solely on that ground, if they possessed an acknowledged currency in general intercourse.” His son notes that “the Thesaurus is an aid not only in the choice of appropriate forms of expression, but in the rejection of those which are unfit; and that a vulgar phrase may often furnish a convenient clue to the group of classic synonyms among which it is placed.” There seem to be several assumptions operating in these sentences that cannot easily be teased apart: if “acknowledged currency” and “unfitness” may embrace the same words and phrases, one wonders who can determine proper “tact” and “taste”, and by what means.
The Thesaurus currently offers a number of vulgarisms and offensive terms, but by no means all. Certain words (related, say, to excreta, drunkenness and genitalia) are sometimes glossed as “slang”, “informal”, or “offensive”: nowhere is the system by which these discriminations are arrived at explained, though “offensive” seems to correspond to words used specifically as insults; the line between slang and informality is harder to discern. In fact, nowhere is it mentioned that such a system is even employed, and the meaning of the abbreviations (sl., inf., derog.) must be inferred by the reader. They are only irregularly applied, and also seem arbitrary in their particular applications.
Of the absence of other “offensive” terminology, one might note the peculiarity by which most racist epithets are absent, despite their “acknowledged currency”, except for the survival of “nigger in the woodpile”, in two entries, evidently considered as inoffensive in each. Englishman is found under “native”: the range of other “natives” specifically offered include the Irish, Welsh and Scots, as well as Muscovites (but not Russians), Parisians and New Yorkers (but not Frenchmen or Americans). This tips one off that “foreigner” might offer some of the other terms, but though it offers some derogatory and slang expressions for members of the United Kingdom (when perceived as foreign, one supposes, though the logic is not impeccable), and a couple of unpleasant words for blacks and Italians, there are so many nations missing, and so many “derogs”, “sls”, “offensives” and “infs” absent, that one wonders what combination of idleness, caution and tact governed the various editors in this area. “Jock” and “taffy”, who were only “native” in 1987, are also “foreign” now, perhaps reflecting the confusion after the recent devolutionary measures: but the index has not been changed accordingly, so one only finds this out by accident.
These are, of course, areas in which no lexicographer is able to find a perfect solution, though the silences and sudden judgments of the Thesaurus are an unsatisfactory approach to the issue. On neologism, the means by which such a book might hope to persuade its existing readers to upgrade their old copy, one enters a similar minefield. The OED and Chambers have criteria for inclusion of new words; these criteria change over time, but one assumes that they are consistently applied at a given period. The editors of the Thesaurus, building on the magpie gatherings of Roget himself, have given no account of their decisions in the area of neologism, though the marketing material (as with any new dictionary) will trumpet the newcomers, offering a partial snapshot of areas of linguistic innovation that, generally, reflect areas of cultural attention and social change. Thus, my 1987 Thesaurus was pleased to note “acid rain, AIDS, genetic fingerprinting, nuclear winter … arbitrageur, creative accounting, insider trading, junk-bond, rate-capping, secondary picketing … baby boomer, bag lady, Sloane Ranger, streetwise and yuppy”. (Don’t the British 1980s come flooding back to you on reading that list?)
One problem with novelty, of course, is that no one has had time to analyse it. Concepts arise that may not last; and concepts may not be perfectly understood. Yuppy, now almost routinely “derog.” in its application, was cheerfully picked up by the 1987 Thesaurus and placed under the category of “Newness” under the subhead “modernist”: semantically it was rubbing shoulders with “bright young thing”, “trendy” and “baby-boomer”, and sharing a wider paragraph with “futurist” and “avant-garde”—rather than, say, “upstart” or “parvenu”. The “newness”, surely, is only in the coinage rather than in the connotation—youthful professional mobility and fast-rising income—which might have seen the word happily installed under “Wealth”, with the subdivision “rich person” offering it the company of “jeunesse dorée” or “new rich” or “nouveau riche” or, less snottily, “self-made man”. Semantically, “Newness” seems the wrong place, especially since the phenomenon was not wholly modern: but there yuppy remains in the latest edition.
The 150th Anniversary Edition offers, inter alia, “genetically modified organisms”, “Frankenstein food”, “air rage”, “noise pollution”, “economic migrant”, “election fatigue”, “health tourist”; Americanisms include a surely belated “lethal injection” and “wake up and smell the coffee”. Its other innovation is to add a dash of Dictionary of Quotations in offering the background to a now familiar phrase, and a dash of Brewer in giving a partial but sizeable list of collective nouns, phobias and suchlike. These strike me as ornamental additions, on the whole; though not unwelcome.
There are two proper roles for a thesaurus. One is mnemonic; the word on the tip of your tongue or at the back of your mind can be located because you know what it means and can thus find it via related terms. (This does rely on the taxonomic structure of the book being efficient: taking yuppy as our example, we can see that, in the rare cases the Thesaurus gets it wrong, it poses a problem for such a reader.) The other role is not to find a fancier word than the one you have; the words that cluster around your own word are less likely to be delicious synonyms than careful variations within a semantic field, each taking you further from what you were saying. No, the other role is, as Dr Roget first thought, a chance to think laterally; by becoming aware of the related semantic opportunities, via homonyms, homophones, synonyms and antonyms, and to note the embedded metaphors of our language, and the discriminations that our culture has required, invented, or insisted upon. It inadvertently teaches us something about who we are; it tells us what our “snow” is. In this way, in its encyclopedic sprawl, its generous ambitions, its decisions and indecisions, and in its constant reminders of its failures and of our own, it might perhaps be classed itself as a vast and extraordinary experimental poem.