【正文】 to change so m uch as a wor d, the argument runs, is to change the meaning as wel l. This aust er e doctr ine has a cer tai n theoreti cal appeal … . Yet at the same t ime this doctr ine l eads t o the alt oget her counteri ntuit ive concl usi on that there can be no such thi ng as st yl e, or t hat styl e is sim ply a par t of content. To put t he pr oblem more concretel y, t he idea of st yl e i mpli es t hat t he words on page mi ght have been di f er ent, or dif ferentl y ar anged, wit hout a cor responding di f er ence in subst ance. (―Gener ati ve Gramm ar s and the Concept of Li terar y Style‖, 1964) To back up his argument that there ar e dif fer ent ways of saying the sam e thi ng, Ohmann of fers the f ol lowi ng par aphr ases of ― Af ter dinner, the senat or made a speech‖: When di nner was over , the senat or m ade a speech. A speech was m ade by t he senator aft er di nner . The senator made a post prandial or ati on. And poi nt s out t hat t hese are var iants of t he or igi nal i n a sense whi ch is not tr ue of , say, ―Columbus was brave‖ or ―Col um bus was nauti cal .‖ The di f er ences among ( 1) (3) are chief ly gr ammati cal 。 we shoul d search f or som e si gnif icance, which we m ay cal l st yli st ic value, in the wr it er’ s choi ce t o express hi s/her sense in t his r at her t han that way. The above not ion of style as ―dress of thought ‖ or as manner of expr ession‖ consi st s in t he asumpti on that there is some basi c sense that can be pr eser ved i n dif fer ent r enderi ngs of wor ds or sentence str uct ur es. This is not li kel y to be chal enged i n ever yday uses of language. But i n li ter at ure, par ti cul arl y in poetr y, par aphr asing bees problemat ic. For example, the m et aphor in ―Com e, seel ing ni ght , / Scarf t he tender eye of pit if ul day‖ ( Macbet h, I I . i . 46 47) denies us a paraphrase in ei ther a li teral sense or a hidden meaning. Any par aphr ase woul d devoi d it of i ts r ichness of impl icati ons t hat i nduces us to f ind inter pr etati ons beyond the meanings capt ured by par aphrasing. Such a m et aphor, as Terence Hawkes says, ―is not f ancif ul em br oidery of t he facts. It i s a way of exper ienci ng the f act s .‖ ( Metaphor, 1972) Lit er ary devices, i n addi ti on to metaphor , such as ir ony, ambi gui ty, pun, and even i mages, poet ry. Wi th del iberate considerati on of this fact, som e theori st s, especial ly t he New Cr i tics, reject the f orm meaning dichotomy and they tend t o see sense and styl e as one t hing, as Wim sat t asser ts: It i s har dly necesary t o adduce pr oof t hat t he doct ri ne of i dent if y of st yle and meaning i s today f ir mly establi shed. The doct ri ne is, I take it , one [ emphasi s mine] fr om which a modern theori st can har dly escape, or har dly wi shes t o. ( The Prose Style of Samuel Johnson, 1941. ) It i s to be not ed that he emphasis upon t he ar ti sti c integr ity and inviolabil it y of their wor ks is echoed not only i n poets but al so in m any pr ose wri ter s, and we can f ind an art iculat ion i n Tolstoy’s wor ds: ―This i s indeed one of t he si gnif icant f act s about a t rue wor k of ar t—t hat i ts cont ent i n i ts enti rety can be expressed onl y by i t sel f . ‖ Cr it cs hol ding such as idea about style t end to l ook at a work of f icti on as a verbal ar ti fact. They bel ieve t hat in such a verbal art ifact t here can be no separati on of the author’ s creati on of the plot , character, social and m oral li fe, fr om t he language i n whi ch they are port r ayed. As Davi d Lodge put s it : ―The noveli st ’s m edi um i s language: what ever he does, qua novel ist, he does i n and thr ough language, Lodge i s ready to see no dif fer ence between t he ki nd of choi ce a wri ter makes in cal ing a character ―dar khair ed‖ or ―f air ,‖ since al the choices a wr it er m akes are a mat er of language. Lodge al so argues that her e i s no esent ial dif fer ence between poet ry and pr ose and t hat t he fol lowi ng te s appl y t o bot h: I t is i mpossi bl e t o par aphr ase l it erary wri ti ng。 I t is i mpossibl e to tr anslat e a li ter ar y wor k。 and t he grammat ical, r at her t han lexical, aspect of styl e is t he one on whi ch Ohm ann concent rates. Thus i n t he analysis of a wr i ter’s style i n a wor k of f icti on, we should study what t he wri ter has wri t en against the backgr ound of what he /she mi ght have wr it ten。 of tr ansit ive or i nt ransit ive verb const ruct ions) ? Ar e there any unusual order ings (i niti al adverbi al s, f ronti ng of object or plement, etc) ? Do speci al ki nds of clause constr uct ion occur (such as t hose wi th pr epar ator y it or there) ? Noun phrases: Are they rel at ively si mple or plex? Wher e does t he plexit y li e (i n pr em odi fi cat ion by adjecti ves, nouns, et c, or i n post modif icati on by preposi ti on by preposi ti onal phr ases, relat ive clauses, et c)? Ver b phrases: Are there any signif icant depart ures f rom the use of the simpl e past tense? For example, not ice occur rences and f unct ions of the pr esent t ense, of the progr essive aspect, of the perf ect aspect , of modal auxi li ar ies. ot her phr ase t ypes: Is t her e anyt hing t o be sai d about other phrases t ypes, such as preposi ti onal phrases, adver b phrases, adj ect ive phrases? Wor d classes: Having alr eady consi dered maj or wor d classes, we may consi der m inor word cl ases (eg f unct ional wor ds) , such as pr eposi t i ons, conj unct ions, pr onouns, deter miner s, auxi li ari es, int er jecti ons. Ar e par ti cul ar words of these types used f or par ticul ar ef ect ( eg demonst rati ves such as this and t hat , negati ves such as not , not hing)? Gener al : Note whet her any gener al t ypes of gramm at ical constr uct ion ar e used t o special