It's not easy to be at a loss for words when it comes to writing about the new fifth edition of Roget's International Thesaurus. It's big -- um, quick, page 187, category 257, paragraph 17 -- and voluminous, capacious, generous, ample, copious, broad, wide, extensive, expansive, comprehensive. And if more is needed, there's category 158, paragraph 10 ("spacious") and its followers. What's more, the thesaurus has taken a new look at the language and found thousands of new terms that are clearly crying for apposite words in the pursuit of lively exposition.
That's what a thesaurus is all about, a dictionary of synonyms for the person who needs a broad palette to color expressions. Roget's Fifth, for those who prefer unqualified statistics, weighs 4 1/2 pounds, has 1,141 pages and contains more than 325,000 words sorted into 1,073 categories describing just about every aspect of the human condition.
It was put together with the inestimable help of computers, which then supplied the data to the ultimate computer, the human brain, in this case the one belonging to Robert L. Chapman, editor, scholar in medieval English studies and lexicographer of modern slang.
For those who work with words, the advent of another edition of such works as Webster's New International Dictionary or the Oxford English Dictionary is an occasion for an outpouring of strong opinion, either hailing linguistic metamorphosis or deploring an assault on established standards. Roget's, a dictionary of not definitions but synonyms and antonyms, never seems to become the focus of such academic brouhaha, and one suspects that is because the hailers and deplorers have found it an important consultant in fashioning their diatribes.
It's not that this new $16.95 thesaurus, the scion of four forebears since Roget's International first went to press in 1886, is free of nits to pick. HarperCollins, the publisher, unleashed Mr. Chapman, a plain-spoken, white-haired man with beard to match, to have his way with the thesaurus, and he has, indeed, made waves, or at least undulations and surges.
He has added 38 new word categories, such as "substance abuse," "tennis," "biology," "automobile racing," "the environment," "bodily development" and "computer science." Although the thesaurus has its share of new words, it is the organization of them that Chapman believes transforms his work from mere this-word-for-that into a true reference book that sparks ideas.
Unlike the facile thesaurus that is content to give synonyms only, like "ebon" for "black," Roget's International follows the concept of Peter Mark Roget (1779-1869), an English scientist and medical man who kept word lists but classified them into his ideas of conceptual orderliness.
In its most recent incarnation, the thesaurus has a 375-page index of all the words. The index refers the reader to various pages in which the word appears in different contexts.
"Net," for instance, has a score of index listings that place it in sections like "webbing," "snare," "arrest" "weave" and "hockey" to help the quester home in on the appropriate category.
"I take a certain satisfaction in being faithful to Dr. Roget's format," Mr. Chapman said. "I feel that the alphabetical version of thesaurus is a sort of slap-- job. They don't serve at all for Roget's purpose, which, as he put it, is to deal with these groups of words as instruments of thought. The alphabetical way, what shall I say, what's a good term for it?" There was a pause, astonishing from a man who had just compiled a thesaurus. "It's just the meat without the potatoes."
Mr. Chapman has retained the categories but has radically reorganized them to conform to his view of existence, from birth through language, human society, occupations, sports and right into humanity's latest achievements, science and technology. Within the lists (but not always in the index) are found new words: "yuppie," "AIDS," "ecosystem," "infotainment."
Where does the thesaurus compiler go to find words? Everywhere he can.
"John Sinclair at Birmingham University in England had a lot of language, we call it corpus, 30 million words, that's not separate words, there ain't that many, but continuous words of English texts, the kind of stuff linguists like to have, an enormous sampling," Mr. Chapman said.
From AT&T; Bell Laboratories' linguistic researchers in Murray Hill, N.J., he got "60 million words, eight years of Associated Press tapes, keyboarded books, proceedings of the Canadian Parliament for 10 years."
"They take anything they can get," he said. "I think they're interested in building computers that talk."
"Which words of this large computer file to use?" he said "That was most laborious, I had to decide what I wanted, all of those individual decisions. You can't do it with a machine, it needs some human intervention. The computer in the mind is better than anything else yet invented."
A word's longevity figured in his calculations.
"If you think some terms are going to be used for a few more years, and assuming that in the future there is about a 15-year interval between editions, put them in; you're not risking very much," he said.
"It's been pointed out that I had 'Trump' in, in the list of 'rich man' and I was asked if I wouldn't like to have that one out because he's not that rich anymore. No, I think he's a symbolic figure. Same with Scud missile that I put in; it had a brief notoriety and maybe I'd take it back, but there it is and it's not hurting anybody."