T.K. Pratt's Dictionary of Prince Edward Island English (DPEIE), first published in 1988, now reprinted in paperback, is a work of some importance, which, despite flaws, makes a contribution to our sense of a distinctive regional strain in Canadian language and culture.
Pratt's work is intended to take a place among a "larger system" of current dialect dictionaries and continuing general records such as the Oxford English Dictionary (OED) and Webster's New International Dictionary-these works together providing, effectively, an atlas of English around the world. Within the "larger system", the OED and Joseph Wright's English Dialect Dictionaries were the two starting-points for serious treatments of the various forms of English. Other regional and dialect dictionaries followed, with the development of synchronic linguistics in the middle of the twentieth century, among them the Scottish National Dictionary, the Australian National Dictionary, the Dictionary of South African English, and the Dictionary of Caribbean Usage. The Gage Canadian Dictionary attempted to offer a standard for this country. However, it was only with the Dictionary of Newfoundland English (1982), probably still the most admired work of its kind in the world, that serious regional or dialect lexicography took hold in Canada.
The Dictionary of Prince Edward Island English is vastly superior to some other Canadian regional dictionaries, such as Look What the Cat Drug in: Miramichi Dictionary, South Shore Phrase Book, Wetcoast Words: A Dictionary of British Columbia Words & Phrases. These are little more than glossaries and they revel in their quaintness. Pratt's undertaking is of another order. He has given the language of P.E.I. an intensive scholarly exploration, but one is left to wonder just what he has accomplished.
His method is outlined in the introduction. Consider this logic: "Standard language is normally quite uniform within a nation. .It is tempting to think of, or perhaps even diagram, the relationship between standard Canadian English and its regional dialects as a central circle with smaller circles clustered around it." It turns out that the smaller circles have an east-west relation. How could a standard be uniform throughout the country, but obtain only within the centre? Pratt retreats from this diagram in favour of a continuum metaphor, but an underlying confusion is evident. To approach non-standard words he would need a clear view of what is meant by "standard", and how, very specifically, the standard relates to the regional dialect. But he cannot do better than this claim:
".isolating the dialect of any region, such as Prince Edward Island, sometimes requires arbitrary measures."
While there is much of interest in its lexicon, P.E.I. is not a place where extraordinary linguistic developments have occurred. Pratt's methods are a tacit admission that this is so. The editors of the Dictionary of Newfoundland English, for example, sought out and found in great numbers words that entered the language there, appeared first or solely in books about Newfoundland, survived there after disappearing elsewhere, developed special forms or senses, or had a higher or more general degree of use there.
Pratt, on the other hand, defines his subject privatively as "non-standard words as used, or once used, on Prince Edward Island". And his dictionary contains large numbers or words which though not "standard", entered the language somewhere else and have not undergone any particular evolution in P.E.I. His tracking of the migrations of words is a real and indubitable service, but not one of great significance.
The corpus base, the body of research, upon which the dictionary is built, seems oddly haphazard. He offers the first and last citations "on hand", suggesting that entries are rather casually assembled. He quotes Holmes to Watson: "You know my method. It is founded on the observance of unconsidered trifles." However charming, this does not quite excuse a poorly supplied and poorly documented evidence corpus. Oral examples come from anonymous "informants", linguistic snitches, as it were. Whereas the Dictionary of Newfoundland English provides a lengthy list of names and locations for its oral sources, and the Dictionary of Caribbean Usage has a strictly defined tapescript corpus, DPEIE merely notes that "behind each word is an average of seven sources."
Though the introduction and the handling of evidence are unsatisfactory from a scholarly perspective, Pratt has found and preserved some truly fascinating words. One can only rejoice in "dreep", an intransitive verb to which Pratt gives the elegant definition, "Usually of rain, to fall in drops, to drip." We learn that a "gridley grinder" is a severe storm. "Stenchel" is "a mixture of molasses and water, used especially on porridge." A "tweeker" is "a small child." A "clart" is a "dirty, untidy person." A "cow's breakfast" is "a large straw hat"-was ever millinery better named? Such words are worth anyone's attention, and in this respect DPEIE can be recommended to readers for pure pleasure. There is a good deal of beauty in this book.
In his preface, Pratt quotes the advice given to him by Canada's greatest dictionary-maker, George Story: "Do not listen to the counsels of perfection." If dictionaries are not perfectible, they must be at least improvable, and there are aspects of this work that ought to be better. Nonetheless, we are in Pratt's debt for producing a worthy and interesting book.

Richard Greene is an assistant professor of English at the University of Toronto. Deirdre Greene, a lecturer at Memorial University in Newfoundland, served as a lexicographer on The New Shorter Oxford English Dictionary.