to remove the difficulties of those who doubt.

It has been judged that this could only be done effectually by a work systematically arranged; for though the cuftomary usage of any fingle word may be shown even by a dictionary *, yet the whole weight of analogy on every fide can only be displayed by the arrangement of fimilar examples in regular claffes. This is the method here employed; and though the Work may not perhaps contain very many obfervations, which fome enquirer or other has not preoccupied, yet it is hoped that the copioufnefs of its matter altogether, and the clearness of its arrangement, will render it more fatisfactory, and more fit for general ufe, than any other treatise of the kind. The brevity and generalnefs of the rules will probably be found favourable to the memory, and the methodical

*No flight is here intended to a gentleman, who, with a very laudable industry, has compiled a dictionary of pronunciation.

claffing

claffing of the exceptions will prevent them from impeding the application of

the rules.

It is evident upon the flighteft obfervation that the language which we fpeak, though long experience has proved it to be no mean inftrument in the hands of genius or of fcience, is much more perplexed than might be wished with wild and inconsistent irregularities. Hence it is that foreigners acquire it with difficulty, and that few even among ourfelves fpeak it with that degree of purity which may justly be expected from those who are uttering their mother tongue. Its grammatical anomalies belong not, as has been faid, to the prefent undertaking, but if they are more numerous than they should be, it is certain that the irregularities which perplex the pronunciation of it are ftill more troublesome and difgraceful this arifes from the very great imperfectness of our literal notation. What correfpondence there ought properly to be between the written form

of

of a language and its vocal founds, may be shown by a general confideration of the fubject of literal notation; what deviation from these rules of propriety our language exhibits, a very flight inspection of this Treatife will demonftrate.

The purpose of literai notation is to agency convey to the mind, by the of the eye, that which living fpeech communicates by means of the ear*: it is, as it has often been expressed, to render founds vifible. As there is not any natural connection between forms and founds, this combination must be originally the work of arbitrary affignment, and previously to any compact for this purpose, any character may ftand for any found. Yet even in arbitrary appointment, if we would avoid confufion, we muft fubmit to certain rules and to render a fyftem of

* Hic enim ufus eft literarum ut cuftodiant voces, et velut depofitum reddant legentibus. Quintil. Inftit. Orator, I. 3.

literal

literal notation completely perfect, the following circumftances are required: 1. That every articulate found to be expreffed fhould have its own fixed and indifputable reprefentative. 2. That a character appropriated to one found, fhould never be employed to represent another*.

The former of these rules is calculated to prevent a defective notation, the latter a confufed one. By deviation from them it is that orthography becomes imperfect, and orthoepy in confequence doubtful. The ftrict observance of them

*In ftrictnefs, therefore, it is an imperfection that the long and the short sounds of the fame vowels fhould be reprefented by the fame mark: however nearly they may be allied, it is neceffary that they should be kept diftinct, as the confounding them is attended with confiderable inconvenience. Accordingly in the most perfect language that is known (the Greek) an approach to the rigorous obfervance of this rule is made; but the contrary imperfection may well enough be tolerated in a language, if the rules for distinguishing the length of vowels by their fituation, that is, the rules of quantity, be fufficiently general and simple.

would

would produce a perfect and unbroken analogy in the founds of a language; the neglect occafions various and inconvenient anomalies *. Perhaps a language cannot eafily be found, in which these rules are more neglected than they are in English; for not only have we founds which have no diftinct and definite representative, but are fometimes expreffed by one character, and fometimes by another †, but almost every letter in our alphabet

* Redundance might also be confidered as a fault in literal notation; but it is not attended with any inconvenience, while it is kept within due bounds, and these it will not often exceed, as men are not used to invent without occafion. That and s fhould reprefent the fame found, is not inconvenient; but that c and t fhould ever depart from their own powers to affume that of s, is a grofs imperfection. The diftinction of small and capital letters, reprefenting the fame founds, is so far from being a faulty redundance, that it is almost an indispensable requifite.

+ As that found of A, generally called its open found, which is fometimes expreffed by the letter itself, as in chaff, dance, &c. (fee the Lift, p. 4.) and sometimes by au, as in aunt, laugh, &c. (fee pp. 12. and 52.).

Alfo