OBS. 5.—Most of our English grammars have more rules of syntax than are needed, and yet are very deficient in such as are needed. To say, as some do, that articles, adjectives, and participles, agree with nouns, is to teach Greek or Latin syntax, and not English. To throw, as Nutting does, the whole syntax of adverbs into a remark on such a rule of agreement, is to choose disorder for its own sake. To say, with Frost, Hall, Smith, Perley, Kirkham, Sanborn, Rand, and others, "The nominative case governs the verb in number and person," and again, "A verb must agree with its nominative case in number and person," is to confound the meaning of government and agreement, to say the same thing in different words, and to leave the subject of a verb still without a rule: for rules of government are applicable only to the words governed, and nothing ever agrees with that which governs it.[325] To say, with Murray and others, "Participles have the same government as the verbs from which they are derived," is to say nothing by which either verbs or participles may be parsed, or any of their errors corrected: those many grammarians, therefore, who make this their only rule for participles, leave them all without any syntax. To say, with Murray, Alger, and others, "Adverbs, though they have no government of case, tense, &c., require an appropriate situation in the sentence," is to squander words at random, and leave the important question unanswered, "To what do adverbs relate?" To say again, with the same gentlemen, "Conjunctions connect the same moods and tenses of verbs, and cases of nouns and pronouns," is to put an ungrammatical, obscure, and useless assertion, in the place of an important rule. To say merely, "Prepositions govern the objective case," is to rest all the syntax of prepositions on a rule that never applies to them, but which is meant only for one of the constructions of the objective case. To say, as many do, "Interjections require the objective case of a pronoun of the first person after them, and the nominative case of the second," is to tell what is utterly false as the words stand, and by no means true in the sense which the authors intend. Finally, to suppose, with Murray, that, "the Interjection does not require a distinct, appropriate rule," is in admirable keeping with all the foregoing quotations, and especially with his notion of what it does require; namely, "the objective case of the first person:" but who dares deny that the following exclamation is good English?
"O wretched we! why were we hurried down This lubric and adulterate age!"—Dryden.
OBS. 6.—The truth of any doctrine in science, can be nothing else than its conformity to facts, or to the nature of things; and chiefly by what he knows of the things themselves, must any one judge of what others say concerning them. Erroneous or inadequate views, confused or inconsistent statements, are the peculiar property of those who advance them; they have, in reality, no relationship to science itself, because they originate in ignorance; but all science is knowledge—it is knowledge methodized. What general rules are requisite for the syntactical parsing of the several parts of speech in English, may be seen at once by any one who will consider for a moment the usual construction of each. The correction of false syntax, in its various forms, will require more—yes, five times as many; but such of these as answer only the latter purpose, are, I think, better reserved for notes under the principal rules. The doctrines which I conceive most worthy to form the leading canons of our syntax, are those which are expressed in the twenty-four rules above. If other authors prefer more, or fewer, or different principles for their chief rules, I must suppose, it is because they have studied the subject less. Biased, as we may be, both by our knowledge and by our ignorance, it is easy for men to differ respecting matters of expediency; but that clearness, order, and consistency, are both expedient, and requisite, in didactic compositions, is what none can doubt.
OBS. 7.—Those English grammarians who tell us, as above, that syntax is divided into parts, or included under a certain number of heads, have almost universally contradicted themselves by treating the subject without any regard to such a division; and, at the same time, not a few have somehow been led into the gross error of supposing broad principles of concord or government where no such things exist. For example, they have invented general RULES like these: "The adjective agrees with its noun in number, case, and gender."—Bingham's English Gram., p. 40. "Interjections govern the nominative case, and sometimes the objective: as, 'O thou! alas me!'"—Ib., p. 43. "Adjectives agree with their nouns in number."—Wilbur and Livingston's Gram., p. 22. "Participles agree with their nouns in number."—Ib., p. 23. "Every adjective agrees in number with some substantive expressed or understood."— Hiley's Gram., Rule 8th, p. 77. "The article THE agrees with nouns in either number: as, The wood, the woods."—Bucke's Classical Grammar of the English Language, p. 84. "O! oh! ah! require the accusative case of a pronoun in the first person after them: as 'Ah me!' But when the second person is used, it requires a nominative case: as, 'O thou!'"—Ib., p. 87. "Two or more Nominatives in the singular number, connected by the Conjunction or, nor, EITHER, NEITHER, govern a singular Verb. But Pronouns singular, of different persons, joined by or, EITHER, nor, NEITHER, govern a plural Verb."—Ib., p. 94. "One Nominative frequently governs many Verbs."—Ib., p. 95. "Participles are sometimes governed by the article."—Murray's Gram., 8vo, p. 192. "An adverb, an adjective, or a participle, may involve in itself the force of a preposition, and govern the objective case."—Nutting's Gram., p. 99. "The nominative case governs the verb." [326]—Greenleaf's Gram., p. 32; Kirkham's, 176; and others. "The nominative case comes before the verb."—Bingham's Gram., p. 38; Wilbur and Livingston's, 23. "The Verb TO BE, always governs a Nominative, unless it be of the Infinitive Mood."—Buchanan's Syntax, p. 94. "A verb in the infinitive mood may be governed by a verb, noun, adjective, participle, or pronoun."—Kirkham's Gram., p. 187. Or, (as a substitute for the foregoing rule,) say, according to this author: "A verb in the infinitive mood, refers to some noun or pronoun, as its subject or actor."—Ib., p. 188. Now what does he know of English grammar, who supposes any of these rules to be worthy of the place which they hold, or have held, in the halls of instruction?
OBS. 8.—It is a very common fault with the compilers of English grammars, to join together in the same rule the syntax of different parts of speech, uniting laws that must ever be applied separately in parsing. For example: "RULE XI. Articles and adjectives relate to nouns expressed or understood; and the adjectives this, that, one, two, must agree in number with the nouns to which they relate."—Comly's Gram., p. 87. Now, in parsing an article, why should the learner have to tell all this story about adjectives? Such a mode of expressing the rule, is certainly in bad taste; and, after all, the syntax of adjectives is not here comprised, for they often relate to pronouns. "RULE III. Every adjective and participle belongs to some noun or pronoun expressed or understood."—Frost's El. of Gram., p. 44. Here a compiler who in his etymology supposes participles to be verbs, allows them no other construction than that of adjectives. His rule implicitly denies that they can either be parts of their verbs in the formation of tenses, or be governed by prepositions in the character of gerunds. To suppose that a noun may govern the objective case, is both absurd in itself, and contrary to all authority; yet, among his forty-nine rules, this author has the following: "RULE XXV. A participial noun is sometimes governed by a preposition, and may govern an objective case; as, 'George is too fond of wasting time in trifles.'"—Frost's El. of Gram., p. 47. Here again is the fault of which I am speaking, two rules in one; and this fault is combined with an other still worse. Wasting is a participle, governed by of; and time is a noun, governed by wasting. The latter is a declinable word, and found in the objective case; the former is indeclinable, and found in no case. It is an error to suppose that cases are the only things which are susceptible of being governed; nor is the brief rule, "Prepositions govern the objective case," so very clear a maxim as never to be misapprehended. If the learner infer from it, that all prepositions must necessarily govern the objective case, or that the objective case is always governed by a preposition, he will be led into a great mistake.
OBS. 9.—This error of crowding things together, is still more conspicuous in the following examples: "RULE IV. Every article, adjective, and participle, must qualify some noun, or pronoun, either expressed or understood."—Nutting's Gram., p. 94. "RULE IX. The objective case is governed by a transitive verb or a preposition, usually coming before it."—Ib., p. 98. Here an author who separates participles from verbs, has attempted first to compress the entire syntax of three different parts of speech into one short rule; and, secondly, to embrace all the forms of dependence, incident to objective nouns and pronouns, in an other as short. This brevity is a poor exchange for the order and distribution which it prevents—especially as none of its objects are here reached. Articles do not relate to pronouns, unless the obsolete phrase the which is to be revived;[327] participles have other constructions than those which adjectives admit; there are exceptions to the rules which tie articles to nouns, and adjectives to nouns or pronouns; and the objective case may not only be governed by a participle, but may be put in apposition with an other objective. The objective case in English usually stands for the Latin genitive, dative, accusative, and ablative; hence any rule that shall embrace the whole construction of this one case, will be the sole counterpart to four fifths of all the rules in any code of Latin syntax. For I imagine the construction of these four oblique cases, will be found to occupy at least that proportion of the syntactical rules and notes in any Latin grammar that can be found. Such rules, however, are often placed under false or equivocal titles;[328] as if they contained the construction of the governing words, rather than that of the governed. And this latter error, again, has been transferred to most of our English grammars, to the exclusion of any rule for the proper construction of participles, of adverbs, of conjunctions, of prepositions, or of interjections. See the syntax of Murray and his copyists, whose treatment of these parts of speech is noticed in the fifth observation above.
OBS. 10.—It is doubtless most convenient, that, in all rules for the construction of cases, nouns and pronouns be taken together; because the very same doctrines apply equally well to both, and a case is as distinct a thing in the mind, as a part of speech. This method, therefore, I have myself pursued; and it has indeed the authority of all grammarians—not excepting those who violate its principles by adopting two special rules for the relative pronoun, which are not needed. These special rules, which I shall notice again hereafter, may be seen in Murray's Rule 6th, which is double, and contains them both. The most complex rule that I have admitted, is that which embraces the government of objectives by verbs and participles. The regimen by verbs, and the regimen by participles, may not improperly be reckoned distinct principles; but the near alliance of participles to their verbs, seems to be a sufficient reason for preferring one rule to two, in this instance.
OBS. 11.—An other common fault in the treatment of this part of grammar, is the practice of making many of the rules double, or even triple, in their form. Of L. Murray's twenty-two rules, for instance, there are six which severally consist of two distinct paragraphs; and one is composed of three such parts, with examples under each. Five others, though simple in their form, are complex in their doctrine, and liable to the objections which have been urged above against this characteristic. These twelve, therefore, I either reject entirely from my catalogue, or divide and simplify to fit them for their purpose. In short, by comparing the twenty-two rules which were adopted by this popular grammarian, with the twenty-four which are given in this work, the reader may see, that twelve of the former have pleased me too little to have any place at all among the latter, and that none of the remaining ten have been thought worthy to be copied without considerable alteration. Nor are the rules which I adopt, more nearly coincident with those of any other writer. I do not proffer to the schools the second-hand instructions of a mere compiler. In his twenty-two rules, independently of their examples, Hurray has used six hundred and seventeen words, thus giving an average of twenty-eight to each rule; whereas in the twenty-four rules which are presented above, the words are but four hundred and thirty-six, making the average less than nineteen. And yet I have not only divided some of his propositions and extended others, but, by rejecting what was useless or erroneous, and filling up the deficiencies which mark his code, I have delivered twice the amount of doctrine in two thirds of the space, and furnished eleven important rules which are not contained in his grammar. Thus much, in this place, to those who so frequently ask, "Wherein does your book differ from Murray's?"
OBS. 12.—Of all the systems of syntax, or of grammar, which it has been my fortune to examine, a book which was first published by Robinson and Franklin of New York in 1839, a fair-looking duodecimo volume of 384 pages, under the brief but rather ostentatious title, "THE GRAMMAR of the English Language" is, I think, the most faulty,—the most remarkable for the magnitude, multitude, and variety, of its strange errors, inconsistencies, and defects. This singular performance is the work of Oliver B. Peirce, an itinerant lecturer on grammar, who dates his preface at "Rome, N. Y., December 29th, 1838." Its leading characteristic is boastful innovation; it being fall of acknowledged "contempt for the works of other writers."—P. 379. It lays "claim to singularity" as a merit, and boasts of a new thing under the sun—"in a theory RADICALLY NEW, a Grammar of the English Language; something which I believe," says the author, "has NEVER BEFORE BEEN FOUND."—P. 9. The old scholastic notion, that because Custom is the arbitress of speech, novelty is excluded from grammar, this hopeful reformer thoroughly condemns; "repudiating this sentiment to the full extent of it," (ib.) and "writing his theory as though he had never seen a book, entitled an English Grammar."—Ib. And, for all the ends of good learning, it would have been as well or better, if he never had. His passion for novelty has led him not only to abandon or misapply, in an unprecedented degree, the usual terms of the art, but to disregard in many instances its most unquestionable principles, universal as well as particular. His parts of speech are the following ten: "Names, Substitutes, Asserters, Adnames, Modifiers, Relatives, Connectives, Interrogatives, Repliers, and Exclamations."—The Gram., p. 20. His names are nouns; his substitutes are pronouns, and any adjectives whose nouns are not expressed; his asserters are verbs and participles, though the latter assert nothing; his adnames are articles, adjectives whose nouns or pronouns are expressed, and adverbs that relate to adjectives; his modifiers are such adverbs as "modify the sense or sound of a whole sentence;" his relatives are prepositions, some of which govern no object; his connectives are conjunctions, with certain adverbs and phrases; his interrogatives and repliers are new parts of speech, very lamely explained; his exclamations are interjections, and "phrases used independently; as, O hapless choice!"—The Gram., p. 22. In parsing, he finds a world of "accommodatives;" as, "John is more than five years older than William."—Ib. p. 202. Here he calls the whole phrase "more than five years" "a secondary adname" i. e., adjective. But, in the phrase, "more than five years afterwards," he would call the same words "a secondary modifier;" i. e., adverb.—Ib., p. 203. And, in the phrase, "more than five years before the war," he would call them "a secondary relative;" i. e., preposition.—Ib., p. 204. And so of other phrases innumerable. His cases are five, two of which are new, "the Independent" and "the Twofold case." His "independent case" is sometimes the nominative in form, as "thou" and "she;" (p. 62;) sometimes the objective, as, "me" and "him;" (p. 62 and p. 199;) sometimes erroneously supposed to be the subject of a finite verb; while his nominative is sometimes as erroneously said to have no verb. His code of syntax has two sorts of rules, Analytical and Synthetical. The former are professedly seventeen in number; but, many of them consisting of two, three, or four distinct parts, their real number is more properly thirty-four. The latter are reckoned forty-five; but if we count their separate parts, they are fifty-six: and these with the others make ninety. I shall not particularize their faults. All of them are whimsically conceived and badly written. In short, had the author artfully designed to turn English grammar into a subject of contempt and ridicule, by as ugly a caricature of it as he could possibly invent, he could never have hit the mark more exactly than he has done in this "new theory"—this rash production, on which he so sincerely prides himself. Alone as he is, in well-nigh all his opinions, behold how prettily he talks of "COMMON SENSE, the only sure foundation of any theory!" and says, "On this imperishable foundation—this rock of eternal endurance—I rear my superstructure, the edifice of scientific truth, the temple of Grammatical consistency!"—Peirce's Preface, p. 7.
OBS. 13.—For the teaching of different languages, it has been thought very desirable to have "a Series of grammars, Greek, Latin, English, &c., all, so far as general principles are concerned, upon the same plan, and as nearly in the same words as the genius of the languages would permit."—See Bullions's Principles of E. Gram., 2d Ed., pp. iv and vi. This scheme necessarily demands a minute comparison not only of the several languages themselves, but also of the various grammars in which their principles, whether general or particular, are developed. For by no other means can it be ascertained to what extent uniformity of this kind will be either profitable to the learner, or consistent with truth. Some books have been published, which, it is pretended, are thus accommodated to one an other, and to the languages of which they treat. But, in view of the fact, that the Latin or the Greek grammars now extant, (to say nothing of the French, Spanish, and others,) are almost as various and as faulty as the English, I am apprehensive that this is a desideratum not soon to be realized,—a design more plausible in the prospectus, than feasible in the attempt. At any rate, the grammars of different languages must needs differ as much as do the languages themselves, otherwise some of their principles will of course be false; and we have already seen that the nonobservance of this has been a fruitful source of error in respect to English syntax. The achievement, however, is not altogether impossible, if a man of competent learning will devote to it a sufficient degree of labour. But the mere revising or altering of some one grammar in each language, can scarcely amount to any thing more than a pretence of improvement. Waiving the pettiness of compiling upon the basis of an other man's compilation, the foundation of a good grammar for any language, must be both deeper and broader than all the works which Professor Bullions has selected to build upon: for the Greek, than Dr. Moor's "Elementa Linguæ, Græcæ;" for the Latin, than Dr. Adam's "Rudiments of Latin and English Grammar;" for the English, than Murray's "English Grammar," or Lennie's "Principles of English Grammar;" which last work, in fact, the learned gentleman preferred, though he pretends to have mended the code of Murray. But, certainly, Lennie never supposed himself a copyist of Murray; nor was he to much extent an imitator of him, either in method or in style.
OBS. 14.—We have, then, in this new American form of "The Principles of English Grammar," Lennie's very compact little book, altered, enlarged, and bearing on its title-page (which is otherwise in the very words of Lennie) an other author's name, and, in its early editions, the false and self-accusing inscription, "(ON THE PLAN OF MURRAY'S GRAMMAR.)" And this work, claiming to have been approved "by the most competent judges," now challenges the praise not only of being "better adapted to the use of academies and schools than any yet published" but of so presenting "the rules and principles of general grammar, as that they may apply to, and be in perfect harmony with, the grammars of the dead languages"— Recommendations, p. iv. These are admirable professions for a critical author to publish; especially, as every rule or principle of General Grammar, condemning as it must whoever violates it, cannot but "be in perfect harmony with" every thing that is true. In this model for all grammars, Latin, Greek, &c., the doctrines of punctuation, of abbreviations, and of capital letters, and also sections on the rhetorical divisions of a discourse, the different kinds of composition, the different kinds of prose composition, and the different kinds of poetry, are made parts of the Syntax; while his hints for correct and elegant writing, and his section on the composition of letters and themes, which other writers suppose to belong rather to syntax, are here subjoined as parts of Prosody. In the exercises for parsing appended to his Etymology, the Doctor furnishes twenty-five Rules of Syntax, which, he says, "are not intended to be committed to memory, but to be used as directions to the beginner in parsing the exercises under them."—E. Gram., p. 75. Then, for his syntax proper, he copies from Lennie, with some alterations, thirty-four other rules, nine of which are double, and all are jumbled together by both authors, without any regard to the distinction of concord and government, so common in the grammars of the dead languages, and even, so far as I can discover, without any principle of arrangement whatever. They profess indeed to have placed those rules first, which are eaisest [sic—KTH] to learn, and oftenest to be applied; but the syntax of articles, which even on this principle should have formed the first of the series, is placed by Lennie as the thirty-fourth rule, and by his amender as the thirty-second. To all this complexity the latter adds twenty-two Special Rules, with an abundance of "Notes" "Observations" and "Remarks" distinguished by these titles, on some principle which no one but the author can understand. Lastly, his method of syntactical parsing is not only mixed up with etymological questions and answers, but his directions for it, with their exemplification, are perplexingly at variance with his own specimen of the performance. See his book, pages 131 and 133. So much for this grand scheme.
OBS. 15.—Strictures like the foregoing, did they not involve the defence of grammar itself, so as to bear upon interests more important than the success or failure of an elementary book, might well be withheld through motives of charity, economy, and peace. There is many a grammar now extant, concerning which a truly critical reader may know more at first sight, than ever did he that made it. What such a reader will be inclined to rate beneath criticism, an other perhaps will confidently pronounce above it. If my remarks are just, let the one approve them for the other's sake. For what becomes of the teaching of grammar, when that which is received as the most excellent method, must be exempted from censure by reason of its utter worthlessness? And what becomes of Universal Syntax, when the imperfect systems of the Latin and Greek grammars, in stead of being amended, are modelled to the grossest faults of what is worthless in our own?[329]
OBS. 16.—What arrangement of Latin or Greek syntax may be best in itself, I am not now concerned to show. Lily did not divide his, as others have divided the subject since; but first stated briefly his three concords, and then proceeded to what he called the construction of the several parts of speech, taking them in their order. The three concords of Lily are the following: (1.) Of the Nominative and Verb; to which the accusative before an infinitive, and the collective noun with a plural verb, are reckoned exceptions; while the agreement of a verb or pronoun with two or more nouns, is referred to the figure syllepsis. (2.) Of the Substantive and Adjective; under which the agreement of participles, and of some pronouns, is placed in the form of a note. (3.) Of the Relative and Antecedent; after which the two special rules for the cases of relatives are given as underparts. Dr. Adam divided his syntax into two parts; of Simple Sentences, and of Compound Sentences. His three concords are the following: (1.) Of one Substantive with an Other; which construction is placed by Lily and many others among the figures of syntax, and is called apposition. (2.) Of an Adjective with a Substantive; under which principle, we are told to take adjective pronouns and participles. (3.) Of a Verb with a Nominative; under which, the collective noun with a verb of either number, is noticed in an observation. The construction of relatives, of conjunctions, of comparatives, and of words put absolute, this author reserves for the second part of his syntax; and the agreement of plural verbs or pronouns with joint nominatives or antecedents, which Ruddiman places in an observation on his four concords, is here absurdly reckoned a part of the construction of conjunctions. Various divisions and subdivisions of the Latin syntax, with special dispositions of some particular principles of it, may be seen in the elaborate grammars of Despauter, Prat, Ruddiman, Grant, and other writers. And here it may be proper to observe, that, the mixing of syntax with etymology, after the manner of Ingersoll, Kirkham, R. W. Green, R. C. Smith, Sanborn, Felton, Hazen, Parkhurst, Parker and Fox, Weld, and others, is a modern innovation, pernicious to both; either topic being sufficiently comprehensive, and sufficiently difficult, when they are treated separately; and each having, in some instances, employed the pens of able writers almost to the exclusion of the other.
OBS. 17.—The syntax of any language must needs conform to the peculiarities of its etymology, and also be consistent with itself; for all will expect better things of a scholar, than to lay down positions in one part of his grammar, that are irreconcilable with what he has stated in an other. The English language, having few inflections, has also few concords or agreements, and still fewer governments. Articles, adjectives, and participles, which in many other languages agree with their nouns in gender, number, and case, have usually, in English, no modifications in which they can agree with their nouns. Yet Lowth says, "The adjective in English, having no variation of gender and number, cannot but agree with the substantive in these respects."—Short Introd. to Gram., p. 86. What then is the agreement of words? Can it be anything else than their similarity in some common property or modification? And is it not obvious, that no two things in nature can at all agree, or be alike, except in some quality or accident which belongs to each of them? Yet how often have Murray and others, as well as Lowth, forgotten this! To give one instance out of many: "Gender has respect only to the third person singular of the pronouns, he, she, it."—Murray, J. Peirce, Flint, Lyon, Bacon, Russell, Fisk, Maltby, Alger, Miller, Merchant, Kirkham, and other careless copyists. Yet, according to these same gentlemen, "Gender is the distinction of nouns, with regard to sex;" and, "Pronouns must always agree with their antecedents, and the nouns for which they stand, in gender." Now, not one of these three careless assertions can possibly be reconciled with either of the others!
OBS. 18.—Government has respect only to nouns, pronouns, verbs, participles, and prepositions; the other five parts of speech neither govern nor are governed. The governing words may be either nouns, or verbs, or participles, or prepositions; the words governed are either nouns, or pronouns, or verbs, or participles. In parsing, the learner must remember that the rules of government are not to be applied to the governing words, but to those which are governed; and which, for the sake of brevity, are often technically named after the particular form or modification assumed; as, possessives, objectives, infinitives, gerundives. These are the only things in English, that can properly be said to be subject to government; and these are always so, in their own names; unless we except such infinitives as stand in the place of nominatives. Gerundives are participles governed by prepositions; but, there being little or no occasion to distinguish these from other participles, we seldom use this name. The Latin Gerund differs from a participle, and the English Gerundive differs from a participial noun. The participial noun may be the subject or the object of a verb, or may govern the possessive case before it, like any other noun; but the true English gerundive, being essentially a participle, and governing an object after it, like any other participle, is itself governed only by a preposition. At least, this is its usual and allowed construction, and no other is acknowledged to be indisputably right.
OBS. 19.—The simple Relations of words in English, (or those several uses of the parts of speech which we may refer to this head,) are the following nine: (1.) Of Articles to nouns, by Rule 1st; (2.) Of Nominatives to verbs, by Rule 2d; (3.) Of Nominatives absolute or independent, by Rule 8th; (4.) Of Adjectives to nouns or pronouns, by Rule 9th; (5.) Of Participles to nouns or pronouns, by Rule 20th; (6.) Of Adverbs to verbs, participles, &c., by Rule 21st; (7.) Of Conjunctions as connecting words, phrases, or sentences, by Rule 22nd; (8.) Of Prepositions as showing the relations of things, by Rule 23d; (9.) Of Interjections as being used independently, by Rule 24th.
OBS. 20.—The syntactical Agreements in English, though actually much fewer than those which occur in Latin, Greek, or French, may easily be so reckoned as to amount to double, or even triple, the number usually spoken of by the old grammarians. The twenty-four rules above, embrace the following ten heads, which may not improperly be taken for so many distinct concords: (1.) Of a Noun or Pronoun in direct apposition with another, by Rule 3d; (2.) Of a Noun or Pronoun after a verb or participle not transitive, by Rule 6th; (3.) Of a Pronoun with its antecedent, by Rule 10th; (4.) Of a Pronoun with a collective noun, by Rule 11th; (5.) Of a Pronoun with joint antecedents, by Rule 12th; (6.) Of a Pronoun with disjunct antecedents, by Rule 13th; (7.) Of a Verb with its nominative, by Rule 14th; (8.) Of a Verb with a collective noun, by Rule 15th; (9.) Of a Verb with joint nominatives, by Rule 16th; (10.) Of a Verb with disjunct nominatives, by Rule 17th. To these may be added two other special concords, less common and less important, which will be explained in notes under the rules: (11.) Of one Verb with an other, in mood, tense, and form, when two are connected so as to agree with the same nominative; (12.) Of Adjectives that imply unity or plurality, with their nouns, in number.
OBS. 21.—Again, by a different mode of reckoning them, the concords or the general principles of agreement, in our language, may be made to be only three or four; and some of these much less general, than they are in other languages: (1.) Words in apposition agree in case, according to Rule 3d; of which principle, Rule 6th may be considered a modification. (2.) Pronouns agree, with their nouns, in person, number, and gender, according to Rule 10th; of which principle, Rules 11th, 12th, and 13th, may be reckoned modifications. (3.) Verbs agree with their nominatives, in person and number, according to Rule 14th; of which principle Rules 15th, 16th, and 17th, and the occasional agreement of one verb with an other, may be esteemed mere modifications. (4.) Some adjectives agree with their nouns in number. These make up the twelve concords above enumerated.
OBS. 22.—The rules of Government in the best Latin grammars are about sixty; and these are usually distributed (though not very properly) under three heads; "1. Of Nouns. 2. Of Verbs. 3. Of Words indeclinable."— Grant's Lat. Gram., p. 170. "Regimen est triplex: 1. Nominum. 2. Verborum. 3. Vocum indeclinabilium."—Ruddiman's Gram., p. 138. This division of the subject brings all the titles of the rules wrong. For example, if the rule be, "Active verbs govern the accusative case," this is not properly "the government of verbs" but rather the government of the accusative by verbs. At least, such titles are equivocal, and likely to mislead the learner. The governments in English are only seven, and these are expressed, perhaps with sufficient distinctness, in six of the foregoing rules: (1.) Of Possessives by nouns, in Rule 4th; (2.) Of Objectives by verbs, in Rule 5th; (3.) Of Objectives by participles, in Rule 5th; (4.) Of Objectives by prepositions, in Rule 7th; (5.) Of Infinitives by the preposition to, in Rule 18th; (6.) Of Infinitives by the verbs bid, dare, &c., in Rule 19th; (7.) Of Participles by prepositions, in Rule 20th.
OBS. 23.—The Arrangement of words, (which will be sufficiently treated of in the observations hereafter to be made on the several rules of construction,) is an important part of syntax, in which not only the beauty but the propriety of language is intimately concerned, and to which particular attention should therefore be paid in composition. But it is to be remembered, that the mere collocation of words in a sentence never affects the method of parsing them: on the contrary, the same words, however placed, are always to be parsed in precisely the same way, so long as they express precisely the same meaning. In order to show that we have parsed any part of an inverted or difficult sentence rightly, we are at liberty to declare the meaning by any arrangement which will make the construction more obvious, provided we retain both the sense and all the words unaltered; but to drop or alter any word, is to pervert the text under pretence of resolving it, and to make a mockery of parsing. Grammar rightly learned, enables one to understand both the sense and the construction of whatsoever is rightly written; and he who reads what he does not understand, reads to little purpose. With great indignity to the muses, several pretenders to grammar have foolishly taught, that, "In parsing poetry, in order to come at the meaning of the author, the learner will find it necessary to transpose his language."—Kirkham's Gram., p. 166. See also the books of Merchant, Wilcox, O. B. Peirce, Hull, Smith, Felton, and others, to the same effect. To what purpose can he transpose the words of a sentence, who does not first see what they mean, and how to explain or parse them as they stand?
OBS. 24.—Errors innumerable have been introduced into the common modes of parsing, through a false notion of what constitutes a simple sentence. Lowth, Adam, Murray, Gould, Smith, Ingersoll, Comly, Lennie, Hiley, Bullions, Wells, and many others, say, "A simple sentence has in it but one subject, and one finite verb: as, 'Life is short.'"—L. Murray's Gram., p. 141. In accordance with this assertion, some assume, that, "Every nominative has its own verb expressed or understood;" and that, "Every verb (except in the infinitive mood and participle) has its own nominative expressed or understood."—Bullions's E. Gram., p. 87. The adopters of these dogmas, of course think it right to supply a nominative whenever they do not find a separate one expressed for every finite verb, and a verb whenever they do not find a separate one expressed for every nominative. This mode of interpretation not only precludes the agreement of a verb with two or more nominatives, so as to render nugatory two of the most important rules of these very gentlemen's syntax; but, what is worse, it perverts many a plain, simple, and perfect sentence, to a form which its author did not choose, and a meaning which he never intended. Suppose, for example, the text to be, "A good constitution and good laws make good subjects."—Webster's Essays, p. 152. Does not the verb make agree with constitution and laws, taken conjointly? and is it not a perversion of the sentence to interpret it otherwise? Away then with all this needless subaudition! But while we thus deny that there can be a true ellipsis of what is not necessary to the construction, it is not to be denied that there are true ellipses, and in some men's style very many. The assumption of O. B. Peirce, that no correct sentence is elliptical, and his impracticable project of a grammar founded on this principle, are among the grossest of possible absurdities.
OBS. 25.—Dr. Wilson says, "There may be several subjects to the same verb, several verbs to the same subject, or several objects to the same verb, and the sentence be simple. But when the sentence remains simple, the same verb must be differently affected by its several adjuncts, or the sense liable to be altered by a separation. If the verb or the subject be affected in the same manner, or the sentence is resolvable into more, it is compounded. Thus, 'Violet, indigo, blue, green, yellow, orange, and red, mixed in due proportion, produce white,' is a simple sentence, for the subject is indivisible. But, 'Violet, indigo, blue, green, yellow, orange, and red, are refrangible rays of light,' is a compound sentence, and may be separated into seven."—Essay on Gram., p. 186. The propriety of the distinction here made, is at least questionable; and I incline to consider the second example a simple sentence, as well as the first; because what the writer calls a separation into seven, involves a change of are to is, and of rays to ray, as well as a sevenfold repetition of this altered predicate, "is a refrangible ray of light." But the parser, in interpreting the words of others, and expounding the construction of what is written, has no right to alter anything in this manner. Nor do I admit that he has a right to insert or repeat anything needlessly; for the nature of a sentence, or the syntax of some of its words, may often be altered without change of the sense, or of any word for an other: as, "'A wall seven feet high;' that is, 'A wall which is seven feet high.'"—Hiley's Gram., p. 109. "'He spoke and acted prudently;' that is, 'He spoke prudently, and he acted prudently.'"—Ibid. '"He spoke and acted wisely;' that is, 'He spoke wisely, and he acted wisely.'"—Murray's Gram., p. 219; Alger's, 70: R. C. Smith's, 183; Weld's, 192; and others. By this notion of ellipsis, the connexion or joint relation of words is destroyed.
OBS. 26.—Dr. Adam, who thought the division of sentences into simple and compound, of sufficient importance to be made the basis of a general division of syntax into two parts, has defined a simple sentence to be, "that which has but one nominative, and one finite verb;" and a compound sentence, "that which has more than one nominative, or one finite verb." And of the latter he gives the following erroneous and self-contradictory account: "A compound sentence is made up of two or more simple sentences or phrases, and is commonly called a Period. The parts of which a compound sentence consists, are called Members or Clauses. In every compound sentence there are either several subjects and one attribute, or several attributes and one subject, or both several subjects and several attributes; that is, there are either several nominatives applied to the same verb, or several verbs applied to the same nominative, or both. Every verb marks a judgment or attribute, and every attribute must have a subject. There must, therefore, be in every sentence or period, as many propositions as there are verbs of a finite mode. Sentences are compounded by means of relatives and conjunctions; as, Happy is the man who loveth religion, and practiseth virtue."—Adam's Gram., p. 202; Gould's, 199; and others.
OBS. 27.—Now if every compound sentence consists of such parts, members, or clauses, as are in themselves sentences, either simple or compound, either elliptical or complete; it is plain, in the first place, that the term "phrases" is misapplied above, because a phrase is properly only a part of some simple sentence. And if "a simple sentence is that which has but one nominative and one finite verb," and "a compound sentence is made up of two or more simple sentences," it follows, since "all sentences are either simple or compound," that, in no sentence, can there be "either several nominatives applied to the same verb, or several verbs applied to the same nominative." What, therefore, this author regarded as the characteristic of all compound sentences, is, according to his own previous positions, utterly impossible to any sentence. Nor is it less repugnant to his subsequent doctrine, that, "Sentences are compounded by means of relatives and conjunctions;" for, according to his notion, "A conjunction is an indeclinable word, which serves to join sentences together."—Adam's Gram., p. 149. It is assumed, that, "In every sentence there must be a verb and a nominative expressed or understood."—Ib., p. 151. Now if there happen to be two nominatives to one verb, as when it was said, "Even the winds and the sea obey him;" this cannot be anything more than a simple sentence; because one single verb is a thing indivisible, and how can we suppose it to form the most essential part of two different sentences at once?
OBS. 28.—The distinction, or real difference, between those simple sentences in which two or more nominatives or verbs are taken conjointly, and those compound sentences in which there is an ellipsis of some of the nominatives or verbs, is not always easy to be known or fixed; because in many instances, a supposed ellipsis, without at all affecting the sense, may obviously change the construction, and consequently the nature of the sentence. For example: "And they all forsook him, and [they all] fled."—Mark, xiv, 50. Some will say, that the words in brackets are here understood. I may deny it, because they are needless; and nothing needless can form a true ellipsis. To the supplying of useless words, if we admit the principle, there may be no end; and the notion that conjunctions join sentences only, opens a wide door for it. For example: "And that man was perfect and upright, and one that feared God, and eschewed evil."—Job, i, 1. No additional words will make this clause any plainer, and none are really necessary to the construction; yet some grammarians will parse it with the following impletions, or more: "And that man was a perfect man, and he was an upright man, and he was one man that feared God, and that eschewed evil things." It is easy to see how this liberty of interpretation, or of interpolation, will change simple sentences to compound sentences, as well as alter the nature and relation of many particular words; and at the same time, it takes away totally those peculiarities of construction by which Dr. Adam and others would recognize a sentence as being compound. What then? are there not two kinds of sentences? Yes, truly; but these authors are wrong in their notions and definitions of both. Joint nominatives or joint verbs may occur in either; but they belong primarily to some simple sentences, and only for that reason are found in any that are compound. A sentence, too, may possibly be made compound, when a simple one would express the whole meaning as well or better; as, "And [David] smote the Philistines from Geba until thou come to Gazer."—2 Sam., v, 25. Here, if we omit the words in Italics, the sentence will become simple, not elliptical.
THE ANALYZING OF SENTENCES.
To analyze a sentence, is, to resolve it into some species of constituent parts, but most properly into words, its first significant elements, and to point out their several relations and powers in the given connexion.
The component parts of a sentence are members, clauses, phrases, or words. Some sentences, which are short and simple, can only be divided into their words; others, which are long and complex, may be resolved into parts again and again divisible.
Of analysis applicable to sentences, there are several different methods; and, so far as their difference may compatibly aid the application of different principles of the science of grammar, there may be an advantage in the occasional use of each.
FIRST METHOD OF ANALYSIS.
Sentences not simple may be reduced to their constituent members, clauses, or simple sentences; and the means by which these are united, may be shown. Thus:—
EXAMPLE ANALYZED.
"Even the Atheist, who tells us that the universe is self-existent and indestructible—even he, who, instead of seeing the traces of a manifold wisdom in its manifold varieties, sees nothing in them all but the exquisite structures and the lofty dimensions of materialism—even he, who would despoil creation of its God, cannot look upon its golden suns, and their accompanying systems, without the solemn impression of a magnificence that fixes and overpowers him."—DR. CHALMERS, Discourses on Revelation and Astronomy, p. 231.
ANALYSIS.—This is a compound sentence, consisting of three complex members, which are separated by the two dashes. The three members are united in one sentence, by a suspension of the sense at each dash, and by two virtual repetitions of the subject, "Atheist" through the pronoun "he," put in the same case, and representing this noun. The sense mainly intended is not brought out till the period ends. Each of the three members is complex, because each has not only a relative clause, commencing with "who," but also an antecedent word which makes sense with "cannot look," &c. The first of these relative clauses involves also a subordinate, supplementary clause,—"the universe is self-existent and indestructible"—introduced after the verb "tells" by the conjunction "that." The last phrase, "without the solemn impression," &c., which is subjoined by "without" to "cannot look," embraces likewise a subordinate, relative clause,—"that fixes and overpowers him,"—which has two verbs; the whole, antecedent and all, being but an adjunct of an adjunct, yet an essential element of the sentence.
SECOND METHOD OF ANALYSIS.
Simple sentences, or the simple members of compound sentences, may be resolved into their PRINCIPAL and their SUBORDINATE PARTS; the subject, the verb, and the case put after or governed by the verb, being first pointed out as THE PRINCIPAL PARTS; and the other words being then detailed as ADJUNCTS to these, according to THE SENSE, or as adjuncts to adjuncts. Thus:—
EXAMPLE ANALYZED.
"Fear naturally quickens the flight of guilt. Rasselas could not catch the fugitive, with his utmost efforts; but, resolving to weary, by perseverance, him whom he could not surpass in speed, he pressed on till the foot of the mountain stopped his course."—DR. JOHNSON, Rasselas, p. 23.
ANALYSIS.—The first period here is a simple sentence. Its principal parts are—Fear, quickens, flight; Fear being the subject, quickens the verb, and flight the object. Fear has no adjunct; naturally is an adjunct of quickens; the and of guilt are adjuncts of flight. The second period is composed of several clauses, or simple members, united. The first of these is also a simple sentence, having, three principal parts—Rasselas, could catch, and fugitive; the subject, the verb, and its object, in their order. Not is added to could catch, reversing the meaning; the is an adjunct to fugitive; with joins its phrase to could not catch; but his and utmost are adjuncts of efforts. The word but connects the two chief members as parts of one sentence. "Resolving to weary" is an adjunct to the pronoun he, which stands before pressed. "By perseverance," is an adjunct to weary. Him is governed by weary, and is the antecedent to whom. "Whom he could not surpass in speed," is a relative clause, or subordinate simple member, having three principal parts—he, could surpass, and whom. Not and in speed are adjuncts to the verb could surpass. "He pressed on" is an other simple member, or sentence, and the chief clause here used, the others being subjoined to this. Its principal parts are two, he and pressed; the latter taking the particle on as an adjunct, and being intransitive. The words dependent on the nominative he, (to wit, resolving, &c.,) have already been mentioned. Till is a conjunctive adverb of time, connecting the concluding clause to pressed on. "The foot of the mountain stopped his course," is a subordinate clause and simple member, whose principal parts are—the subject foot, the verb stopped, and the object course. The adjuncts of foot are the and of the mountain; the verb in this sentence has no adjunct but course, which is better reckoned a principal word; lastly, his is an adjunct to course, and governed by it.
THIRD METHOD OF ANALYSIS.
Sentences may be partially analyzed by a resolution into their SUBJECTS and their PREDICATES, a method which some late grammarians have borrowed from the logicians; the grammatical subject with its adjuncts, being taken for the logical subject; and the finite verb, which some call the grammatical predicate[330] being, with its subsequent case and the adjuncts of both, denominated the predicate, or the logical predicate. Thus:—
EXAMPLE ANALYZED.
"Such is the emptiness of human enjoyment, that we are always impatient of the present. Attainment is followed by neglect, and possession, by disgust. Few moments are more pleasing than those in which the mind is concerting measures for a new undertaking. From the first hint that wakens the fancy, to the hour of actual execution, all is improvement and progress, triumph and felicity."—DR. JOHNSON, Rambler.
ANALYSIS.—Here the first period is a compound sentence, containing two clauses,—which are connected by that. In the first clause, emptiness is the grammatical subject, and "the emptiness of human enjoyment" is the logical. Is some would call the grammatical predicate, and "Such is," or is such, the logical; but the latter consists, as the majority teach, of "the copula" is, and "the attribute," or "predicate," such. In the second clause, (which explains the import of "Such,") the subject is we; which is unmodified, and in which therefore the logical form and the grammatical coincide and are the same. Are may here be called the grammatical predicate; and "are always impatient of the present," the logical. The second period, too, is a compound sentence, having two clauses, which are connected by and. Attainment is the subject of the former; and, "is followed by neglect" is the predicate. In the latter, possession alone is the subject; and, "[is followed] by disgust," is the predicate; the verb is followed being understood at the comma. The third period, likewise, is a compound, having three parts, with the two connectives than and which. Here we have moments for the first grammatical subject, and Few moments for the logical; then, are for the grammatical predicate, and are more pleasing for the logical: or, if we choose to say so, for "the copula and the attribute." "Than those," is an elliptical member, meaning, "than are those moments," or, "than those moments are pleasing;" both subject and predicate are wholly suppressed, except that those is reckoned a part of the logical subject. In which is an adjunct of is concerting, and serves well to connect the members, because which represents those, i.e. those moments. Mind, or the mind, is the next subject of affirmation; and is concerting, or, "is concerting measures for a new undertaking," is the predicate or matter affirmed. Lastly, the fourth period, like the rest, is compound. The phrases commencing with From and to, describe a period of time, and are adjuncts of the verb is. The former contains a subordinate relative clause, of which that (representing hint) is the subject, and wakens, or wakens the fancy, the predicate. Of the principal clause, the word all, taken as a noun, is the subject, whether grammatical or logical; and "the copula," or "grammatical predicate," is, becomes, with its adjuncts and the nominatives following, the logical predicate.
FOURTH METHOD OF ANALYSIS.
All syntax is founded on the RELATION of words one to an other, and the CONNEXION of clauses and phrases, according to THE SENSE. Hence sentences may be, in some sort, analyzed, and perhaps profitably, by the tracing of such relation or connexion, from link to link, through a series of words, beginning and ending with such as are somewhat remote from each other, yet within the period. Thus:—
EXAMPLES ANALYZED.
1. "Swift would say, 'The thing has not life enough in it to keep it sweet;' Johnson, 'The creature possesses not vitality sufficient to preserve it from putrefaction.'"—MATT. HARRISON, on the English Language, p. 102. ANALYSIS.—What is the general sense of this passage? and what, the chain of connexion between the words Swift and putrefaction? The period is designed to show, that Swift preferred words of Saxon origin; and Johnson, of Latin. It has in contrast two coördinate members, tacitly connected: the verb would say being understood after Johnson, and perhaps also the particle but, after the semicolon. Swift is the subject of would say; and would say introduces the clause after it, as what would be said. The relates to thing; thing is the subject of has; has, which is qualified by not, governs life; life is qualified by the adjective enough, and by the phrase, in it; enough is the prior term of to; to governs keep; keep governs it, which stands for the thing; and it, in lieu of the thing, is qualified by sweet. The chief members are connected either by standing in contrast as members, or by but, understood before Johnson. Johnson is the subject of would say, understood: and this would say, again introduces a clause, as what would be said. The relates to creature; creature is the subject of possesses; possesses, which is qualified by not, governs vitality; vitality is qualified by sufficient; sufficient is the prior term of to; to governs preserve; preserve governs it, and is the prior term of from; and from governs putrefaction.
2. "There is one Being to whom we can look with a perfect conviction of finding that security, which nothing about us can give, and which nothing about us can take away."—GREENWOOD; Wells's School Gram., p. 192.[331]
ANALYSIS.—What is the general structure of this passage? and what, the chain of connexion "between the words away and is?" The period is a complex sentence, having four clauses, all connected together by relatives; the second, by whom, to the first and chief clause, "There is one Being;" the third and the fourth, to the second, by which and which; but the last two, having the same antecedent, security, and being coördinate, are also connected one to the other by and. As to "the chain of connexion," Away relates to can take; can take agrees with its nominative nothing, and governs which; which represents security; security is governed by finding; finding is governed by of; of refers back to conviction; conviction is governed by with; with refers back to can look; can look agrees with we, and is, in sense, the antecedent of to; to governs whom; whom represents Being; and Being is the subject of is.
FIFTH METHOD OF ANALYSIS.
The best and most thorough method of analysis is that of COMPLETE SYNTACTICAL PARSING; a method which, for the sake of order and brevity, should ever be kept free from all mixture of etymological definitions or reasons, but which may be preceded or followed by any of the foregoing schemes of resolution, if the teacher choose to require any such preliminary or subsidiary exposition. This method is fully illustrated in the Twelfth Praxis below.
OBSERVATIONS ON METHODS OF ANALYSIS.
OBS. 1.—The almost infinite variety in the forms of sentences, will sometimes throw difficulty in the way of the analyzer, be his scheme or his skill what it may. The last four or five observations of the preceding series have shown, that the distinction of sentences as simple or compound, which constitutes the chief point of the First Method of Analysis above, is not always plain, even to the learned. The definitions and examples which I have given, will make it generally so; and, where it is otherwise, the question or puzzle, it is presumed, cannot often be of much practical importance. If the difference be not obvious, it can hardly be a momentous error, to mistake a phrase for an elliptical clause, or to call such a clause a phrase.
OBS. 2.—The Second Method above is, I think, easier of application than any of the rest; and, if other analysis than the regular method of parsing seem desirable, this will probably be found as useful as any. There is, in many of our popular grammars, some recognition of the principles of this analysis—some mention of "the principal parts of a sentence," in accordance with what are so called above,—and also, in a few, some succinct account of the parts called "adjuncts;" but there seems to have been no prevalent practice of applying these principles, in any stated or well-digested manner. Lowth, Murray, Alger, W. Allen, Hart, Hiley, Ingersoll, Wells, and others, tell of these "PRINCIPAL PARTS;"—Lowth calling them, "the agent, the attribute, and the object;" (Gram., p. 72;)—Murray, and his copyists, Alger, Ingersoll, and others, calling them, "the subject, the attribute, and the object;"—Hiley and Hart calling them, "the subject or nominative, the attribute or verb, and the object;"—Allen calling them, "the nominative, the verb, and (if the verb is active,) the accusative governed by the verb;" and also saying, "The nominative is sometimes called the subject; the verb, the attribute; and the accusative, the object;"—Wells calling them, "the subject or nominative, the verb, and the object;" and also recognizing the "adjuncts," as a species which "embraces all the words of a simple sentence [,] except the principal parts;"—yet not more than two of them all appearing to have taken any thought, and they but little, about the formal application of their common doctrine. In Allen's English Grammar, which is one of the best, and likewise in Wells's, which is equally prized, this reduction of all connected words, or parts of speech, into "the principal parts" and "the adjuncts," is fully recognized; the adjuncts, too, are discriminated by Allen, as "either primary or secondary," nor are their more particular species or relations overlooked; but I find no method prescribed for the analysis intended, except what Wells adopted in his early editions but has since changed to an other or abandoned, and no other allusion to it by, Allen, than this Note, which, with some appearance of intrusion, is appended to his "Method of Parsing the Infinitive Mood:"—"The pupil may now begin to analyse [analyze] the sentences, by distinguishing the principal words and their adjuncts."—W. Allen's E. Gram., p. 258.
OBS. 3.—These authors in general, and many more, tell us, with some variation of words, that the agent, subject, or nominative, is that of which something is said, affirmed, or denied; that the attribute, verb, or predicate, is that which is said, affirmed, or denied, of the subject; and that the object, accusative, or case sequent, is that which is introduced by the finite verb, or affected by the action affirmed. Lowth says, "In English the nominative case, denoting the agent, usually goes before the verb, or attribution; and the objective case, denoting the object, follows the verb active."—Short Introd., p. 72. Murray copies, but not literally, thus: "The nominative denotes the subject, and usually goes before the verb [,] or attribute; and the word or phrase, denoting the object, follows the verb: as, 'A wise man governs his passions.' Here, a wise man is the subject; governs, the attribute, or thing affirmed; and his passions, the object."—Murray's Octavo, p. 142; Duodecimo, 116. To include thus the adjuncts with their principals, as the logicians do, is here manifestly improper; because it unites what the grammatical analyzer is chiefly concerned to separate, and tends to defeat the main purpose for which "THE PRINCIPAL PARTS" are so named and distinguished.
OBS. 4.—The Third Method of Analysis, described above, is an attempt very briefly to epitomize the chief elements of a great scheme,—to give, in a nutshell, the substance of what our grammarians have borrowed from the logicians, then mixed with something of their own, next amplified with small details, and, in some instances, branched out and extended to enormous bulk and length. Of course, they have not failed to set forth the comparative merits of this scheme in a sufficiently favourable light. The two ingenious gentlemen who seem to have been chiefly instrumental in making it popular, say in their preface, "The rules of syntax contained in this work result directly from the analysis of propositions, and of compound sentences; and for this reason the student should make himself perfectly familiar with the sections relating to subject and predicate, and should be able readily to analyze sentences, whether simple or compound, and to explain their structure and connection. * * * This exercise should always precede the more minute and subsidiary labor of parsing. If the latter be conducted, as it often is, independently of previous analysis, the principal advantage to be derived from the study of language, as an intellectual exercise, will inevitably be lost."—Latin Grammar of Andrews and Stoddard, p. vi. N. Butler, who bestows upon this subject about a dozen duodecimo pages, says in his preface, "The rules for the analysis of sentences, which is a very useful and interesting exercise, have been taken from Andrews' and Stoddard's Latin Grammar, some changes and additions being made."—Butler's Practical Gram., p. iv.[332]
OBS. 5.—Wells, in the early copies of his School Grammar, as has been hinted, adopted a method of analysis similar to the Second one prescribed above; yet referred, even from the first, to "Andrews and Stoddard's Latin Grammar," and to "De Sacy's General Grammar," as if these were authorities for what he then inculcated. Subsequently, he changed his scheme, from that of Parts Principal and Adjuncts, to one of Subjects and Predicates, "either grammatical or logical," also "either simple or compound;"—to one resembling Andrews and Stoddard's, yet differing from it, often, as to what constitutes a "grammatical predicate;"—to one resenbling [sic—KTH] the Third Method above, yet differing from it, (as does Andrews and Stoddard's,) in taking the logical subject and predicate before the grammatical. "The chapter on Analysis," said he then, "has been Revised and enlarged with great care, and will be found to embody all the most important principles on this subject [.] which are contained in the works of De Sacy, Andrews and Stoddard, Kühner, Crosby, and Crane. It is gratifying to observe that the attention of teachers is now so generally directed to this important mode of investigating the structure of our language, in connection with the ordinary exercises of etymological and syntactical parsing."—Wells's School Gram., New Ed., 1850, p. iv.
OBS. 6.—In view of the fact, that Wells's chief mode of sentential analysis had just undergone an almost total metamorphosis, a change plausible perhaps, but of doubtful utility,—that, up to the date of the words just cited, and afterwards, so far and so long as any copies of his early "Thousands" remain in use, the author himself has earnestly directed attention to a method which he now means henceforth to abandon,—in this view, the praise and gratulation expressed above seem singular. If it has been found practicable, to slide "the attention of teachers," and their approbation too, adroitly over from one "important mode of investigating the structure of our language," to an other;—if "it is gratifying to observe," that the direction thus given to public opinion sustains itself so well, and "is so generally" acquiesced in;—if it is proved, that the stereotyped praise of one system of analysis may, without alteration, be so transferred to an other, as to answer the double purpose of commending and superseding;—it is not improbable that the author's next new plates will bear the stamp of yet other "most important principles" of analysis. This process is here recommended to be used "in connection with the ordinary exercises of etymological and syntactical parsing,"—exercises, which, in Wells's Grammar, are generally, and very improperly, commingled; and if, to these, may be profitably conjoined either his present or his former scheme of analysis, it were well, had he somewhere put them together and shown how.
OBS. 7.—But there are other passages of the School Grammar, so little suited to this notion of "connection" that one can hardly believe the word ought to be taken in what seems its only sense. "Advanced classes should attend less to the common Order of Parsing, and more to the Analysis of language."—Wells's Grammar, "3d Thousand," p. 125; "113th Thousand," p. 132. This implies, what is probably true of the etymological exercise, that parsing is more rudimental than the other forms of analysis. It also intimates, what is not so clear, that pupils rightly instructed must advance from the former to the latter, as to something more worthy of their intellectual powers. The passage is used with reference to either form of analysis adopted by the author. So the following comparison, in which Parsing is plainly disparaged, stands permanently at the head of "the chapter on Analysis," to commend first one mode, and then an other: "It is particularly desirable that pupils should pass as early as practicable from the formalities of common PARSING, to the more important exercise of ANALYZING critically the structure of language. The mechanical routine of technical parsing is peculiarly liable to become monotonous and dull, while the practice of explaining the various relations and offices of words in a sentence, is adapted to call the mind of the learner into constant and vigorous action, and can hardly fail of exciting the deepest interest,"—Wells's Gram., 3d Th., p. 181; 113th Th., p. 184.
OBS. 8.—An ill scheme of parsing, or an ill use of a good one, is almost as unlucky in grammar, as an ill method of ciphering, or an ill use of a good one, would be in arithmetic. From the strong contrast cited above, one might suspect that, in selecting, devising, or using, a technical process for the exercising of learners in the principles of etymology and syntax, this author had been less fortunate than the generality of his fellows. Not only is it implied, that parsing is no critical analysis, but even what is set in opposition to the "mechanical routine," may very well serve for a definition of Syntactical Parsing—"the practice of explaining the various relations and offices of words in a sentence!" If this "practice," well ordered, can be at once interesting and profitable to the learner, so may parsing. Nor, after all, is even this author's mode of parsing, defective though it is in several respects, less "important" to the users of his book, or less valued by teachers, than the analysis which he sets above it.
OBS. 9.—S. S. Greene, a public teacher in Boston, who, in answer to a supposed "demand for a more philosophical plan of teaching the English language," has entered in earnest upon the "Analysis of Sentences," having devoted to one method of it more than the space of two hundred duodecimo pages, speaks of analysis and of parsing, thus: "The resolving of a sentence into its elements, or of any complex element into the parts which compose it, is called analysis."—Greene's Analysis, p. 14. "Parsing consists in naming a part of speech, giving its modifications, relation, agreement or dependence, and the rule for its construction. Analysis consists in pointing out the words or groups of words which constitute the elements of a sentence. Analysis should precede parsing."—Ib., p. 26. "A large proportion of the elements of sentences are not single words, but combinations or groups of words. These groups perform the office of the substantive, the adjective, or the adverb, and, in some one of these relations, enter in as the component parts of a sentence. The pupil who learns to determine the elements of a sentence, must, therefore, learn the force of these combinations before he separates them into the single words which compose them. This advantage is wholly lost in the ordinary methods of parsing."—Ib., p. 3.
OBS. 10.—On these passages, it may be remarked in the first place, that the distinction attempted between analysis and parsing is by no means clear, or well drawn. Nor indeed could it be; because parsing is a species of analysis. The first assertion would be just as true as it is now, were the former word substituted for the latter: thus, "The resolving of a sentence into its elements, or of any complex element into the parts which compose it, is called parsing." Next, the "Parsing" spoken of in the second sentence, is Syntactical Parsing only; and, without a limitation of the species, neither this assertion nor the one concerning precedence is sufficiently true. Again, the suggestion, that, "Analysis consists in pointing out the words or groups of words which constitute the elements of a sentence," has nothing distinctive in it; and, without some idea of the author's peculiar system of "elements," previously impressed upon the mind, is scarcely, if at all, intelligible. Lastly, that a pupil must understand a sentence,—or, what is the same thing, "learn the force of the words combined,"—before he can be sure of parsing each word rightly, is a very plain and certain truth; but what "advantage" over parsing this truth gives to the lesser analysis, which deals with "groups," it is not easy to discover. If the author had any clear idea of "this advantage," he has conveyed no such conception to his readers.
OBS. 11.—Greene's Analysis is the most expanded form of the Third Method above.[333] Its nucleus, or germinating kernel, was the old partition of subject and predicate, derived from the art of logic. Its chief principles may be briefly stated thus: Sentences, which are simple, or complex, or compound, are made up of words, phrases, and clauses—three grand classes of elements, called the first, the second, and the third class. From these, each sentence must have two elements; the Subject, or Substantive element, and the Predicate, or Predicative element, which are principal; and a sentence may have five, the subordinates being the Adjective element, the Objective element, and the Adverbial element. The five elements have sundry modifications and subdivisions. Each of the five may, like a sentence, be simple, or complex, or compound; and each may be of any of the three grand classes. The development of this scheme forms a volume, not small. The system is plausible, ingenious, methodical, mostly true, and somewhat elaborate; but it is neither very useful nor very accurate. It seems too much like a great tree, beautiful, symmetrical, and full of leaves, but raised or desired only for fruit, yet bearing little, and some of that little not of good quality, but knurly or bitter. The chief end of a grammar, designed for our tongue, is, to show what is, and what is not, good English. To this end, the system in question does not appear to be well adapted.
OBS. 12.—Dr. Bullions, the projector of the "Series of Grammars, English, Latin, and Greek, all on the same plan," inserted in his Latin Grammar, of 1841, a short sketch of the new analysis by "subjects and predicates," "grammatical and logical," the scheme used by Andrews and Stoddard; but his English Grammar, which appeared in 1834, was too early for this "new and improved method of investigating" language. In his later English Grammar, of 1849, however, paying little regard to sameness of "plan" or conformity of definitions, he carefully devoted to this matter the space of fifteen pages, placing the topic, not injudiciously, in the first part of his syntax, and referring to it thus in his Preface: "The subject of ANALYSIS, wholly omitted in the former work, is here introduced in its proper place; and to an extent in accordance with its importance."— Bullions, Analyt. and Pract. Gram., p. 3.