GRAMMAR. 



treatife on grammar, in tlie EncyclopaeJia Britannica, (ar- 

 ticle II.) "The divifion of words into thofe whicli arc 

 p.gnijiiant of thtmfclvcs, andtliofe which are fignificatit by re- 

 lation, is abfolute nonfenfe, and has been productive of mucli 

 error and myftery in fonie of the moil celebrated treatifi-s on 

 grammar. It is indeed probable, tliat any attempt to elln- 

 blilh a different clafTilication of the parts of fpeech, from 

 tnat which is commonly received, will be found of little utili- 

 ty, either mpraake OT mfpeciilalion. As far as they>;) /wr is 

 concerned, the vulgar divifion feems fufficiently commodi- 

 ous ; for every man who knows any thing, knows wlien he 

 tifes a noun, and when a verb. With refpeft to the latler, 

 not to mention that all the grammarians, from Arillotle to 

 Home Tooke, have differed on the fubject ; it (hould feem 

 to be of more importance, after having afcertained, with 

 preciiion, the nature of each fpecies of words, to dcterniine 

 in what circumftances they differ thaji in what they 

 agree." 



If thefe remarks be juft, any attempt at a philofophi- 

 cal arrangement of the^ parts of laiigiiage mufl be frivolous 

 and ufelefs. But we regard the paragraph as a inanifeft ac- 

 knowledgment that the writer knew of no fuch arrange- 

 ment, and that, as is often the cafe, he endeavours to confole 

 himfelf and his readers, under tlie abfence of this knowledge, 

 with the confideration that if attained it would not be ufeful. 

 We iliall, however, attempt briefly to do what he thinks 

 unneceffary to be done : but what we think not only ufeful 

 and necelTary, but even efTential, In a philofophical trcatife. 

 An accurate diftribution of our words, is but an accurate 

 diftribution ofourideas^ and as in logic we cannot think 

 juftly, unlefs we knovr how to diftinguifh our thoughts; fo 

 we cannot with precifion reafon in grammar, unlefs we have 

 pre\ioufly diftinguilhed the parts of fpeech. The fource of 

 this dlil'nilion muil be fought in the nature of the ideas 

 exprefTed by thofe parts, and the enquiry, how many forts 

 of words there exiil in human language, is merely, when 

 dicefted to its proper objedl, how many forts of ideas there 

 are in human knowledge. 



We have then ideas of things: we have ideas of the 

 properties or attributes of things : we have ideas of the con- 

 nedion of things, i. e. of the arrangement of things, or of 

 their increafe or diminution: finally, v.e have ideas of the 

 relation of things. Nor have we an idea which may not rank 

 with one of thefe claffes. And as language confills of terms 

 «xprefnng ideas, it confills of terms which esprefs either 

 ideas of things or the attributes of thinge, or the ionneelion of 

 things, or the relation of things. The terras exprelTnig 

 things are nouns, including fubflantivts and the perfonal 

 pronouns, which arc only fubjlitults for nouns, and there- 

 fore, in their nature, names of things. Siueet is an attribute 

 of fugar, and fo is to think an attribute of a man ; tojiy of 

 a bird ; to be of a being that exills, and io, moreover, lov- 

 ing, hating, are attributes of fome creatures fufceptible of 

 love and hatred; finally, the circumilances which define 

 things or diftinguifh them from others, may be confidered 

 as attributes or properties of the things fo defined. The 

 terms exprelTing attributes may be called atlrilulives ; and 

 include adjeftives, verbs, adverbs, participles, and defini- 

 tives. Tlie terms expreffing the relations of thiggs are pre- 

 pofitions ; wlule thofe which exprefs the conneftion of things 

 are conjunftions. And thus the moft comprehenfive, and, 

 as we conceive, the mod philofophical divifion of fpeecli is 

 into four parts, nouns, atlritutiru, prepoftions, and eonjunc- 

 tions or connectives ; nor is there any language, liowever few 

 or numerous its terms may be, which is not capable of 

 being dillributed into theic four claffes. As to the inter- 

 jfHion, it mull be regarded as a foiu>d wlijch is eitlicr inar- 



ticulate, or which, as far as it partakes of articulation, 

 partakes of the nature of a noun or of a verb. Having 

 noted this diftribution, we, as not fyllcmatically writing on 

 gramnuu-, fiiall, in our remarks, follow the popular divi- 

 fion. 



Nouns and attiibutivcs, to ufe Mr. Harris's language, 

 are fignificent as principals, and wc do not depart from phi- 

 lofophical accuracy, when we fay that prepofitions and con- 

 neiftives are fignificant by relation. Our notions of thr 

 relations and connections of things arife by the principle of 

 abftrac^ion from our notions of the things themfclvc% fo 

 related or connected If all ideas of things or aclior.» were 

 obliterated from our minds, the ideas of tTieir affociattd re- 

 lations would be entirely obliterated willi them. And as 

 our notions of relation fpring from the objecls fo related, 

 fo language, following the rife and progrefs of ideas, bai 

 derived thofe terms which exprefs relations and conneclious, 

 from the terms cxiircHing the things and operations fo re- 

 lated and connected. In other wordi, prepofitions and coii- 

 junction.s are words which, by affociation, flowed firom nounn 

 and verbs as their real foiirces. If this be correct, what fhalt 

 we fay to Mr. H. Tooke, who affirms prepofitions, con- 

 junctions, and other particles, to be fubflitutes for thofe neccf- 

 fary words, the noun and verb. If thii affertion be put 

 in clear language, it means that the relations of things, or 

 of the operations of tilings, are fubjlitutes for the things fo 

 related ; that is, the relation of caufe and effect is a fubflitute- 

 for caufe and effect; the relation of &;f»nn;'nj is a fabftitute 

 for things which begin ; the relation of injlrumentality is a 

 fubilitute for that which is inftrumental; and the relation 

 of end is a fubilitute for that which ends. Moreover, the 

 feparation or concatenation of two or more alfertions is s 

 fubilitute for the operations fo feparatcd or connected. 

 This is not philofophical reafoning, calculated to throw- 

 light on language, but a fophiflry calculated to confound 

 the moil dillinft ideas which the human mind is capable of 

 forming, and a jargon not to be equalled by any error or 

 contradiction, which Mr. Tooke has io fevcrely cenfured 

 and ridiculed in Mr. Harris. Befides, terms to exprefs the 

 relations and connections of things are not lefs neceffary 

 than thofe which denote things and their operations. A 

 language without conjunctions and prepofitions is not merely 

 a fledge, (to ufe his own words,) which cannot he drawn 

 along as fmoothly, and eafily, and fwiftly, as a carriage 

 with wheels, but is a fledge which has nothing to connect it 

 with the horfe ; nothing to direct it in the way along whicli 

 it Ihould be dragged. The noun and the verb indeed may 

 exprefs a complete propofition; but the inllances in which 

 fuch fimple propofitions occur are very few, compared with 

 thofe more complicated cafes where tlic ule of prepofitions 

 and connectives is abfoluteiy neceffary. The diftribution 

 of fpeech into necelTary words a!id abbreviations, is there, 

 fore, in this point of view, alfo frivolous and erroneous. 



The Origin and Properfiei of Kovni. 



Nouns are the names of things, or figns of thofe 

 ideas by which we come to the knowledge ot things. The 

 term is a corruption of nomen, which is from r>-nu->, while 

 ntime came probably from the Perfian njam, which fecms to 

 be of the lame origin vi'.h the Greek word. Nouns liiav 

 be divided into three dalles: names of imiividuals, names of 

 fpecies or hinds, names of abjtraci ideoi. To tliis may be 

 added, as a fourth clafs, xiie pronoun, which is made the fub- 

 ilitute or reprelentative of all other nouns. But we fhall 

 conhder this in the place ufually afllgned to it by gramma- 

 rians. On each of llie above claffes we beg to make a few 

 brief remarks. 



4 D .; Kan'.cs 



