ftone rampart for that purpose. The area inclosed by the inner wall is about eighty paces long, and twenty-seven broad; and both the inner and outer walls appear, by the ruins, to have been exceeding well vitrified.

This feems to be the hill, of which, under the name of Craig Feterick, an account is given in the last volume of the Philofophical Tranfactions, as noticed in our Review for May. The hill is there represented as having once been a volcano; but Mr. Williams's more accurate investigation feems entirely to overturn this conjecture.

After giving an account of three other fortified hills of the fame kind, namely, Caftle Finlay, and Dun-Evan, in the fhire of Nairn, and of Finaven, in the fhire of Angus, Mr. Williams proceeds to deliver his opinion relative to the manner in which those curious buildings have been erected; in which detail we find fome ingenious obfervations on the progress of the human mind in the invention of arts.

The author next makes fome remarks of the ruins of dry ftone buildings, which are found in many parts of the Highlands, and are uniformly of a conic figure. The area, on the ground within the walls, is from thirty to forty foot diameter. The entrance was always by one low door, and they had a cavity at the bottom, running quite round in the heart of the wall, which is conjectured to have been defigned for keeping provifions. Thofe buildings had a small opening at the top, for admitting light, as well as affording a paffage to the fmoke, which rose from the fire, that is supposed to have burned in the middle of the area.

Notwithstanding the obvious difference between the structure of those buildings and the vitrified forts, Mr. Williams thinks it is not improbable that they belonged to the fame period of time, and were raised by the fame people. In fupport of this conjecture he observes, that the vitrified forts are found only where the rock is of the plum pudding kind, which is easily vitrified; and the conic ftru&tures where the ftones are large, square, and broad-bedded, but could not so easily be rendered subject to vitrification.

To the narrative, is fubjoined a description of Craig-Patrick, by Mr. James Watt, engineer; with a letter to Mr. Williams, from Dr. Black, profeffor of chymistry in the univerfity of Edinburgh, in which this ingenious gentleman concurs with him in opinion, refpecting the manner in which he supposes thofe vitrified forts to have been conftructed. The discoveries made by Mr. Williams are not only highly gratifying to curiofity, confidering them as the fubjectof antiquarian refearches; but afford a ftriking inftance of the extraordinary expedients to which people had recourse in the infancy of arts.

A Letter to John Dunning, Efq. By Mr. Horne. 8vo, Is. 6d. Johnfon.

THE author of this Letter takes occafion, from an expreffion in a precedent, quoted at his trial, to enter into a train of grammatical fpeculations.

The point in debate is thus opened and explained.

• A fuppofed omiffion, in the information againft Lawley, is produced to justify a real omiffion, in the information against me; when indeed there was no omiflion in the precedent. But the averment faid to be omitted, was, not only fubftantially, but iterally made.

The exception taken was, that it was not pofitively averred, that Crooke was indicted, it was only laid, that fhe fciens, that Crooke had been indicted, and was to be tried for forgery, did fo and fo."-That is literally thus: "Crooke had been indicted for forgery" (there is the averment literally made)" she knowing that, did so and so.”

Such, fir, is, in all cafes, the unfufpected construction not only in our own, but in every language in the world, where the conjunction that, or fome equivalent word, is employed. I fpeak confidently, becaufe I know, a priori, that it must be fo; and I have likewife tried it in a great variety of lan. guages, ancient as well as modern, Afiatic as well as European.'

The word that, he thinks, is therefore not to be confidered as a conjunction, but as an article, or a pronoun: and to prove this, he produces, among many others, the following examples: 'I wish you to believe, that I would not wilfully hurt a fly.' 'In this inftance the construction, he says, is to be thus refolved: 'I would not wilfully hurt a fly, I wish you to believe that (affertion).-" Thieves rife by night, that they may cut men's throats."-Resolution : Thieves may cut men's throats; (for) that (purpose) they rife by night.'

He adds:

• This method of refolution takes place in those languages, which have different conjunctions for the fame purpose: for the original of the laft example, where ut is employed, and not the the Latin neuter article quod, will be refolved in the fame

manner.

Ut jugulent homines, furgunt de nocte latrones.'

Through San&tius, who ftruggled fo hard to withdraw quod from among the conjunctions, ftill left ut among them without moleftation, yet is ut no other than the Greek article ori, adopted for this conjunctive purpose by the Latins, and by them originally written uti: the o being changed into u from that propenfity which both the ancient Romans had, and the modern

Italians

Italians ftill have, upon many occafions, to pronounce even their own o like an u...The refolution therefore of the original will be like that of the tranflation:"

Latrones jugulent homines (d) or furgunt de nocte.'

• But how are we to bring out the article that, when two conjunctions come together in this manner?

Have

is If that the king

any way your good deferts forgot, He bids you name your griefs."

Shakef.

The truth of the matter is, that if is merely a verb, the imperative mood of the Gothic and Anglo-Saxon verbs gifan; and in those languages, as well as in the English formerly, this fuppofed conjunction was pronounced and written, as the common imperative gif. Thus, in Ben Johnson's Sad Shepherd, it is written :

"My largeffe

Hath lotted her to be your brother's mistreffe, Gif she can be reclaimed; gif not, his prey." Accordingly our corrupted if has always the fignification of the prefent English imperative give, and no other. So that the refolution of the conftruction, in the inftance produced from Shakespeare, will be as before in the others: The king may have forgotten your good deeds: give that in any way, he bids you name your griefs."

And here, as an additional proof, we may obferve, that whenever the datum, upon which any conclufion depends, is a fentence, the article that, if not expreffed, is understood: as, in the inftance produced above, the poet might have faid,

Gif (that) the can be reclaimed, &c.

For the refolution is: "She can be reclaimed, give that, my largeffe hath lotted her to be your brother's miftieffe: fhe cannot be reclaimed, give that, my largeffe hath lotted her to be your brother's prey."

-We have in English another word, which (though now rather obfolete) ufed frequently to fupply the place of if. As,

An you had an eye behind you, you might fee more detraclion at your heels, than fortunes before you.'

No doubt it will be afked; in this and in all fimilar inflances what is an ?

• I do not know that any perfon has ever attempted to explain it, except Dr. S. Johnfon in his Dictionary. He fays,an is fometimes, in old authors, a contraction of and if.”. Of which he gives a very unlucky inftance from Shakespeare; where both an and if are used in the fame line;

"He cannot flatter, he!

An honeft mind and plain; he muft fpeak truth!
An they will take it,-So. If not, he's plain."

• Where

Where if an was a contraction of and if; an and if should rather change places.

But I can by no means agree with Johnfon's account. A part of one word only, employed to fhew that another word is compounded with it, would indeed be a curious method of contraction: although even this account of it would ferve my purpofe; but the truth will ferve it better: for an is also a verb, and may very well fupply the place of if: it being nothing elfe but the imperative mood of the Anglo Saxon verb anan, which likewife means to give or to grant.

• Nor does an ever (as Johnfon fuppofes) fignify as if; nor is it a contraction of them.

I know indeed that Johnfon produces Addifon's authority

for it.

"My next pretty correfpondent, like Shakespeare's lion in Pyramus and Thifbe, roars an it were any nightingale."

Now if Addifon had fo written, I fhould answer roundly, that he had written falfe English. But he never did fo write. He only quoted it in mirth. And Johnfon, an editor of Shakefpeare, ought to have known and obferved it. And then, inflead of Addison's or even Shakespeare's authority from whom the expreffion is borrowed; he fhould have quoted Bottom's, the weaver: whofe language correfponds with the character Shakefpeare has given him.

"I will aggravate my voice fo (fays Bottom) that I will roar you as gently as any fucking dove: I will roar you an 'twere any nightingale.

Our author, having thus accounted for if and an, afferts that those words, which are called conditional conjunctions, are to be accounted for in all languages, in the fame manner. Not that they muft all mean precifely give and grant, but fomething equivalent: as, be it, fuppofe, allow, permit, fuffer, &c.

Hitherto the doctrine of conjunctions has been the crux grammaticorum. Thefe troublesome words have caufed them infinite labour and perplexity. Yet all their etymologies have been vague and unfatisfactory. Mr. Harris tells us, that a conjunction is a part of fpeech, void of fignification;' and he compares them to cement in a building. Lord Monboddo fays, prepofitions, conjunctions, and fuch like words, are rather the pegs and nails that faften the feveral parts of the language together, than the language itfelf.' Mr. Locke declares himself diffatisfied with all the accounts of them, that he had feen. Sandius refcued qued particularly from the num ber of thefe myfterious conjun&ions. Servius, Scioppius, Voffius, Perizonius, and others, have difplaced and explained many other fuppofed adverbs and conjunctions. Dr. Johnfon fays, the particles are, among all nations, applied with to

6

great

great a latitude, that they are not eafily reducible under any regular scheme of interpretation.' He adds: • I have la-` boured them with diligence, I hope with fuccefs: fuch at leaft as can be expected in a task, which no man, however learned or fagacious, has yet been able to perform.'Our author however has undertaken to perform this task; and has actually reduced our principal conjunctions under a regular scheme of interpretation. As this then appears to be a matter of importance, in the theory of our language, we shall give our readers a fummary view of the most material remarks, in this differtation. The learned author, we hope, will pardon us, if we do not reprefent his opinion, in its fullest extent, when he confiders, that no epitome can be adequate to an original work.

IF is the imperative gif of the Saxon verb gifan *, to give. AN is the imperative an, of anan, to grant. These words may be used mutually and indifferently to fupply each others place. Gif is to be found in all our old writers. G. Douglas almost always ufes gif; once or twice only he has used if; and once he uses gewe for gif. Chaucer commonly uses if; but fometimes jeve, jef, and yf. And it is to be observed, that in Chaucer, and other old writers, the verb to give fuffers the fame variations in the manner of writing it, however used, whether conjun&tively, or otherwife.

Well ought a prieft enfample for to yeve.'

Prol. to Cant. Tales.

Gin is often ufed in our northern counties, and by the Scotch, as we ufe if or an: which they do with equal propriety, and as little corruption: for gin is no other than the participle given, gi'en, gi'n.

UNLESS, Onles, is the imperative of the Saxon onlefan, to difmifs. This word is written by Horne, bishop of Winchester, in the reign of queen. Elizabeth, onles, oneles, onleffe, onelesse: by bishop Gardiner, onles, onlesse.-Les the imperative of lesan, which has the fame meaning as onlefan, is likewife used fometimes by old writers inftead of unless. It is the fame imperative at the end of those words which are called adjectives, fuch as bopeless, motionless, i. e. difmifs hope, difmifs motion.

EKE is the imperative eac of eacan, to add.

YET is the imperative get or gyt, of getan or gytan, to get. STILL is the imperative ftell or steall, of stellan, or feallian, to put. These words may very well fupply each others place, and be indifferently used for the fame purpose.

* For the ease of readers, unacquainted with the Saxon characters, we have taken the liberty, throughout this article, to use English letters in their place.

7

ELSE.