THE ENGLISH LANGUAGE AND ENGLISH GRAMMAR i mssss pi HI IBRARY nlVSRSlTYOF CALIFORNIA THE ENGLISH LANGUAGE AND ENGLISH GRAMMAR AN HISTORICAL STUDY OF THE SOURCES, DEVELOPMENT, AND ANALOGIES OF THE LANGUAGE AND OF THE PRINCIPLES GOVERNING ITS USAGES ILLUSTRATED BY COPIOUS EXAMPLES FROM WRITERS OF ALL PERIODS SAMUEL RAMSEY G. P. PUTNAM'S SONS NEW YORK LONDON 37 WEST TWENTY-THIRD STREET 24 BEDFORD STREET, STRAND &t ^nichtrbotktr ^rws 1892 COPYRIGHT, 1892 BY SAMUEL RAMSEY Electrotyped, Printed, and Bound by Ubc Knickerbocker press, TAcw V?orfe G. P. PUTNAM'S SONS PREFACE. MUCH of what is contained in the following pages was first written for purposes of private instruction, and without any view to publication ; but as one chapter was added after another, it began to be thought that some portions might interest a larger class of readers. There are many persons who would be glad to know more about the English language than can be gained from the formal routine of the public schools, who, nevertheless, are unable to procure and read the great number of valuable works on the subject that have issued from the press within a period of fifty years. Such persons are in a position to appreciate a work taking a somewhat wider view than the common text-books, and presenting some of the more familiar results of modern philology. There are two classes for whom this book is not intended. The first are those who are already familiar with all the results of past labors, and who, therefore, can find nothing here to add to their present ample stores of knowledge, there being no claim to original discovery or invention. The second class are those who neither know or care anything about the history or philology of their native tongue. Between these extremes is the large and important class who already know something and desire to know more. There was a fable of the Rabbins that the first pair of blacksmith's tongs were made during the six days of cre- ation, because, without such primordial instrumentality, no tool could ever have been fashioned. I cannot but think that many have unconsciously imbibed a somewhat similar belief in regard to English grammar. It may not be distinctly taught, but everything tends to impress the learner with a iv Preface. vague idea that the rules laid down in his manual were ordained " in the beginning," and have remained unchanged and unchangeable ever since, and that in the fulness of time the English language was made in obedience to them. The reader of these pages will have an opportunity to become acquainted with the opposite doctrine that language, so far as we are acquainted with it, is a human product, subject like others to evolution and mutation as liable to change as the forms of our garments or our dwellings, and that the office of grammar is not to go before and decree what men shall say, but to follow after and describe what they do say. S. R. ERRATA. Page 19, line 23, for acres, read races. " " " " for Higdon, read Trevisa. 21 " iS,f0r ne spe, read nes fe. " 63 " 6, for extradition, raz Note, There was also a pural, cildra, or cildru. " 242 " 29 if, for fit, gis, tit5, read fet, ges, teS. " 2 93>f or t* 16 initial pa, rm^/-everywhere pa. " 294, line 22, for preceded, r^dt^did not precede. " 324 " 20, for ipsi, readipse. " 347 " 21, for Ir J^, read\\se. " " " 22, /0r Venir j^, r^a^ Venire. " 359 " Zif or grafen, razd' grave. " 364 " 28,y<9r sow, read saw. " 375 " i8,for wollede and shullede, read wolde and sceolde. " 412 " 2g,for levdeyr, read levy r. " 428 " 29, /-v, men funden a r*tf ^2^^ and perfyt ; and the thridde day next d//r*, he fleethe his ze/^y. And so there is no mo briddes of that kynde in #//v/ unto God, be cause that there nys no God but on, and, also, that oure lord aroos fro dethe to lyve the thridde day." PROCLAMATION OF HENRY III., 1258. " Jfenr' thurl godes fultume King on Engleneloande. Lhoauerd on Yrloand'. Duk on Norm* on Aquitain' and eorl on Aniow. Send igretinge to alle hise halde ilarde and ileawede on Hunt- endon 1 sehir' that witen 3 us and thurZ that loandes folk on vre kun- 1 This character (3) here represents a modification of the Anglo-Saxon g much in use from the twelfth to the fourteenth century, with a value varying from jf to gh, the latter no longer recognized in English. The Instability of Language. j eriche. habbeth idon and schullen don in the worthnesse of gode and on vre treowthe. for the freme' of the loande. thur~$, the besi^te of than to foreniseide redesmen, beo stedefcest and ilestinde in alle t hinge abut en cende." ' FROM THE "PETERBOROUGH CHRONICLE," 1150. " And te ear I of Angau ward ded, and his sune Henri toe to the rice. And te cuen of France to-dalde fra the king and sees com to the iunge eorl Henri, and he toe hire to wive, and Peitou mid hire. thafarde he mid micel far d into Engleland and m? castles and te kingferde agenes him mid micel mare ferd, thothwathere fuhtten he ). The passage may be read thus : Henry, through God's help, King in England, Lord in Ireland, Duke in Normandy, in Aquitain and Earl in Anjou, sends greeting to all his subjects, learned and lay, in Huntingdonshire. This know ye well all, that we will and grant that which our counsellors all or the more part of them that be chosen through us and through the landfolk of our kingdom, have done and shall do to the honor of God and our allegiance, for the good of the land, through the determination of the aforesaid counsellors, be established and obeyed in all things forever. 8 The English Language. huntg, and fiscath j and se cyning and thd Hcostan men drincath myran meolc and tha unsp]> Godess hellpe Off J>att Judisskenn follkess lac ]>att Drihhtin wass full cvveme, & mikell hellpe to )>e follc, to laeredd & to laewedd, Biforenn fatt te Laferrd Crist was borenn her to marine." And now I will show you Something, with God's help Of that Jewish people's worship That to the Lord was very acceptable And much help to the people, To learned and to unlearned, Before that the Lord Christ Was born here a man. The other poem, entitled " The Brut," was by a priest named Lasamon (pronunciation uncertain, oftener written The Sources of English. 21 Layamon). It is a mythical history of Britain from the sack of Troy to King Athelstan. Brut is a descendant of ^Eneas, \vho after incredible adventures lands in Britain, to which he gives its name. The poem is an amplified translation of a Norman-French poem of the same name ; itself a transla- tion from the Latin original by Geoffrey of Monmouth. " pa 3et spaec Haengest : Lauerd ich wulle fin iwil : & don al mine dsede : Nu ic wulle biliue : & asfter mire dohter : & sefter ohte mo#nen : and J> u 3if me swa muchel lond : swa wule anes bule hude : feor from aelche castle : ***** Of J>ere hude he kaerf enne Jnvong : ne sfe J> wong noht swiSe braed : }>a al islit wes ]>e ]>ong : a-buten he bilaede : Then yet spoke Hengist : Lord, I will thy pleasure : And do all my deeds : Now I will quickly : And after my daughter : And after brave men : If thou give me so much land : As will a bull's hide : Far from each castle : ***** Of the hide he cut a thong : The thong was not very broad : When slit was all the thong ; And about it spread around : cnihten alre hendest. dri3e her & ouer-al. sefter Jrine raede. sende after mine wiue. J>e me is swa deore. fa bezste of mine cunne. to stonden a mire a3ere hod. aelches weies ouer-sprseden. amidden ane ualde. ***** swiSe smal & swifte long. buten swulc a twines f raed, he wes wunder ane long, muche del of londe." of all knights courtliest, do here and everywhere, after thy counsel, send after my wife, that to me is so dear, the best of my kinsmen, to stand in my holding, each way overspread, in midst of a wood. ***** very small and very long, but such as a thread of twine, it then was wondrous long, a mighty deal of ground. From this time onward translations and imitations of French works became more and more frequent, and have 22 The English Language. never once ceased to this day. In the thirteenth century French had become the literary language of Europe ; and nearly all that was worth reading for amusement was derived from that source. It is not from the Norman conquerors but from seven centuries of contact with French literature that we receive the greater part of the French words in our language. Mr. Kington Oliphant says of the period from 1 220 to 1303: " English was cast aside as something vulgar, and nearly every cultivated writer in our island betook himself to French or Latin ; our tongue almost lost its noble power of compounding, and parted with thousands of old words. A very few translations from French and Latin kept a feeble light burning during these baleful years. In Age III., 1280-1303, English writers transla- ted copiously from the French, though they gave birth to noth- ing original ; they thus stopped the decay of our fast perishing language, and French words in shoals were brought in to supply the place of the English lost." But it is not words alone that we have thus acquired. French examples have influenced our pronunciation, spell- ing, and grammatical and literary forms. The following are a few of the most easily distinguishable features due in whole or in part to French influence : 1. The prevalence of the hissing sound with which our language is reproached the sounds which we represent by sh, ch, and j, the sibilant sound of c, the almost universal ending of the plural in s, the verbal ending in s goes and speaks instead of goeth and speaketh. 2. The loss of the guttural sound represented in Saxon by //, and in later English by gh. 3. The loss of a very useful character j>, and the substitu- tion of two letters (tft) in its place. 4. Ownership expressed by of " the house of the planter," instead of " the planter's house." 5. Comparison of adjectives by more and most "the most beautiful," instead of Carlyle's " beautifullest." 6. The placing of the adjective after the noun, or giving Tfu Sources of English. 23 it a plural form sign manual, letters patent, courts martial, Knights Templars. 7. You instead of thou. 8. The union of a verb and noun drawbridge, cutpurse. They are not very numerous; and the latest coinages know-nothing, push-cart, grip-sack, do not make us wish them more plentiful. 9. The anomalous expressions ; " It is me" " That's him" 10. Rhyme and the modern system of versification. The earliest English poetry depended neither on rhyme, accent, nor measure, but on alliteration, that is, identity of initial sounds. This was natural with the Saxons and Scandinavi- ans, because, as a rule, all words were accented on the first syllable. In an old poem on the deluge, God says to Noah : dalles ther-inne, & ^alkes ful mony, 2?oth ^oskes & <5oures, & wel fcmnden penes ; For I schal ze/aken up a water to ze/asch alle the worlde." Two populations of kindred blood and a common worship, and occupying the same country, could not remain separate and hostile forever. Social and family ties began slowly to draw together Saxon and Norman. The wisdom and vigor of some of the Norman kings, the baseness and imbecility of others the generosity, bravery, and wrongs of Richard, the futile tyranny of John, the splendid victories of Crecy and Poictiers and the sight of two captive kings in London at once diverted the thoughts of men from the question of race, and taught them to sympathize, resist, and feel a com- mon pride together. It was during the fourteenth century that the varied population of England became one people, speaking one language, still easily understood by the intelli- gent reader. In this age of rapid transit and sudden revo- lution we are struck with the slowness of progress a few centuries ago. It was two hundred and ninety-seven years after the Conquest that Edward III., in his anger against France and desire to unite all his subjects against that nation, abolished the use of the French language in legal pleadings and public acts. Still time was necessary to give 24 The English Language. full effect to the law, and the oldest public document in English preserved in Rymer's " Federa" is twenty-three years later. We have seen, too, that English was not admitted into the schools until the reign of Richard II. What, then, was the character of this early English, and of what elements was it composed ? As the great body of the people were Anglo-Saxon, we may safely infer on a very general principle that they furnished the framework of the language. If a people whose principal intercourse is with each other have occasion to borrow words from a second people they will be chiefly names of things, especially of new things for which they could have no native names, just as we have picked up such words as caravan, indigo, chintz, manna, alkali, bamboo, gorilla, jalap, canoe, moccasin. Next they would adopt words expressing actions, especially pro- cesses unknown before, and lastly words expressive of qualities. The little words that express nothing by themselves, but are of wondrous convenience the, he, it, any, what, why, in, to, of, if, and, but, though, yet, the winged words that save labor, the articles and particles that express time, number, relations, and conditions there would be no need to borrow. People are satisfied with what they have already. Of the hundreds of words which our present English has borrowed from extra-European sources, all but three are names, or what grammarians call nouns. Of these three, two shampoo and tattoo are verbs, that is, express actions. The third taboo is an adjective, an expression of quality sacred or devoted to the gods. So of 600 French words found in " Robert of Gloucester," 386 are nouns, 140 verbs, 68 adjec- tives, and 6 of all other kinds. The last six are all made of nouns or nouns and particles ; and only two of them because and piecemeal are still in use. This will give a fair idea of the kind of words, grammatically, that were introduced from French and Latin. Another effect of commingling languages would take the form of loss and not of acquisition. A word, while still re- maining the same word, may have little appendages affixed to the beginning or the end, or it may undergo internal changes. The Sources of English. 25 boy boy-s ox ox-en sing sing-s child children love loved speak spoken This phenomenon is known by the general name of inflexion ; and some languages have much more of it than others. In some a word may have several hundred or even a thousand forms ; and in such cases a knowledge of them forms the greatest part of what is called grammar. English has now only remnants and traces of a system once much more extensive. Both Anglo-Saxon and Norman-French had in- flexional systems much more largely developed, but quite unlike and incapable of combining to any great extent. The greater part was dropped altogether. When people under- stand each other imperfectly they cannot preserve a multi- tude of niceties. The ear catches the essential part of the word, and gives little heed to the ever changing termination. This neglect was the more effective as the accent was gen- erally near the beginning of each word and the termination pronounced indistinctly. The colored race in this country when first emancipated afforded an amusing illustration of this. They could not be got to utter the parts of words that preceded or were far from the accent. Ascription, description, inscription, prescription, were all alike scription and nothing more. It is often questioned which is the best preserved part of a language, and, consequently, the best evidence of ancient relationship, the words or the grammatical forms. This question, I think, does not admit of a general and absolute answer. Much depends on circumstances. The principle might be presented in the form of a supposed case. If people of two races A and B, capable of friendly union, be placed together on an island, a mixed language will result. Suppose the people of A to outnumber those of B ten to one, but the intelligence of B to be ten times the greatest ; and further suppose them so far intermingled that the learned B's talk chiefly with the ignorant A's ; then the grammatical system of A will survive in a simplified form, and words will be bor- 26 The English Language. rowed from B, according to necessity or fancy. Much de- pends on the relative numbers brought into intimate contact. There are instances of small numbers belonging to ancient races, scattered among large populations, who retain more or less of their ancestral words, which they use according to the grammatical system of the country. The Armenians scattered through Asia Minor are said to use native words with a Turkish grammar. A German Jew will say to his wife : " Ich habe noch haiyom lo ge-ackalt. " " I have not eaten anything to-day yet." I distinguish the Hebrew por- tions by italics, but the structure of the sentence is purely German. A very interesting example is furnished by the gypsies, who have a considerable vocabulary of their own, but are too much scattered to maintain a grammatical ays- tern. In Spain their grammar is Spanish. All their verbs are of the first Spanish conjugation, which they follow in all its great extent and complexity. In England they adopt the very simple structure of the English language. The follow- ing is from a song of the English gypsies given by George Borrow : " We jaws to the drab-engro ker, Trin /ivrsevsorih there of drab we lels, And when to the swety back we ivels, We pens we '11 drab the baulo." We goes to the poison-master's house, Three pennyworth there of poison we buys, And when to the folks back we comes, We says we '11 poison the pig. Here are merely single gypsy words in a setting of pure English. The grammar is furnished by the majority ; the words by those who know most. And again, words are gathered by wide intercourse ; grammatical forrns are de- veloped by isolation. People who go round the world in sixty days will not wait to transform a single word into a thousand shapes. Having seen that the words adopted by the English were chiefly significant ones, representing things, actions, and The Sources of English. 2 7 qualities, we may next inquire, for what kind of things, qualities, and actions they found it necessary to borrow words from French and Latin these two being almost in- separably connected. The necessity arose from two causes the loss of native words, and the access of new objects and ideas. We may assume that any language, however limited and threadbare, can, by combining, recombining, and modifying its words, develop expressions for all human thoughts. Yet all cannot do this equally well. We know the unlimited copiousness of modern German ; and Anglo-Saxon, a sister tongue, might, under favorable circumstances, have become equally rich. There was a considerable literature before the Conquest, and, by compounding the native words, writers were able to say all that they had occasion for. They had especially an ample stock of words for representing the emotions and ethical ideas. I do not raise the question here whether those words would always commend themselves to our eyes or ears, but they served the purpose intended. But after the Conquest, literature ceased almost wholly. Only a very few cloistered monks read and wrote. Books and the language of books were forgotten. The great body of the Saxon people were in the condition of a low type of farm laborers ; and it is but a small part of a language that such people have occasion for. Marsh thinks that one half of the language had disappeared before the year 1300.' Successive literary specimens show an ever decreasing native vocabulary, deficiencies being supplied at will from French or Latin. But the condition of the Saxons left them at liberty to preserve a multitude of words belonging to every-day rustic life, which are still heard in every hamlet and rural district where the English tongue is spoken. The names for the family rela- tions and the domestic animals are Saxon. I give here some examples of a more miscellaneous character, merely to show 1 "A careful examination of several letters of Bosworth's Anglo-Saxon Dic- tionary gives, in 2,000 words (including derivatives and compounds, but excluding orthographic variants), 535 which still exist as modern English words." Encyclopaedia Britannica, viii., 390. 28 The English Language. more distinctly the kind of words that serfs and rustic labor- ers were in a condition to preserve : land tree sickle to hew hill grass spade to delve dale hay rake to sow marsh fodder axe to reap field thistle hammer to mow meadow nettle nail to thrash sand bramble saw to winnow day briar loom to live loam thorn oats to bake dung fern wheat to brew furrow dike barley to watch ridge ditch straw to wed wood stile chaff to spin water harrow honey to weave well scythe wax to sew From the Norman-French and the early French romances and songs were naturally derived a multitude of words such as are used by the wealthy and governing classes, relating to government, law, war, hunting, dress, furniture, and amuse- ments. The following are a few of the words introduced before A.D. 1300 : amour conquer homage palfrey armor countess honor park arson court jest parlor ball crown jewel parliament banner dame judgment pavilion baron dress jugglery peerage battle duke lance prison castle empire madam ransom chamberlain enemy mantle renown champion ermine marshal rent chancellor falcon messenger sable charter galley miniver scarlet chess gentleman noble tower chivalry governor palace venison The Sources of English. 29 While the difference here exhibited is very general, we need not expect anything in language to be carried out con- sistently. Knighthood, the very crown and blossom of Nor- man Chivalry, is Saxon, while the flail with which the rustic thrashed his barley, was French-Latin. It has often been pointed out that, while the names of the domestic animals, ox, cow, calf, sheep, swine, are Saxon, their flesh, as an article of food, bears the French names, beef, veal, mutton, and pork, with the apparent implication that the Saxons merely raised their flocks and herds for others to eat. While that may be true in part, I do not think it the imme- diate or the principal reason. No doubt the Normans called the animals, whether alive or dead, by French names, and were most immediately interested in them when brought to the table, but what did the Saxons call them in that state ? If they had any special names, they were probably oxna-flcssc, nedt-flcesc, cealf-flcesc, sceop-fl&sc, swin-flcesc, in analogy with modern German ; but the men who knit together the bones and sinews of the present English gener- ally left out long Saxon compounds, not because they were Saxon, but because they were clumsy. Or if the Saxons had no distinctive terms for the flesh of the animals, the greater was the necessity for preserving the French names and restricting their meaning. We have here, as in so many instances, words of similar signification, but from different sources, preserved and assigned different duties. Did space permit, the condition of the two peoples might be outlined by naming their domestic surroundings. The Norman baron dwelt in a castle, " Hemmed in by battlement and fosse, And many a darksome tower" with its barbican and portcullis, its esplanade, court, chapel, stables, and offices. Its central strength was the donjon keep or dungeon, in which were the cellar and pantry, the parlors, chambers, and closets. The beds were surrounded with cur- tains, and the walls hung with tapestries. There the baron and his guests sat on chairs and dined at a table. 30 The English Language. The Saxon churl had still his house and home, hearth-stone, and roof-tree ; but they were unpretentious. The poor man's dwelling had but two apartments, the Scotch but and ben by-out and by-in. It had neither parlor nor chamber, cellar itor garret, closet nor recess, partition nor ceiling. It had not even a chimney ; its roof was of thatch, and its windows were without glass mere eyes, or openings for the wind. The householder might learn from the Norsemen to put up a loft under the roof, to be reached by a ladder. He had neither chair nor table, but sat on a bench, a stool, or a settle, and ate his meat from a board. Outside might be a wort-yard where potherbs grew the parent of the modern orchard, and near by were the barn, the byre, and the sty. As Latin was especially employed for the graver purposes of religion, philosophy, and diplomacy, the words taken directly from it, and not passed through a French filter, would naturally have a character of dignity all their own. Words like abstract, belligerent, conscience, desolate, eternal, formula, genius, hereditary, inviolate, jurist, lunar, millennium, nominalist, omniscient, perpetual, will sufficiently illustrate the character I refer to. It is observed, too, that when we have words from other sources for a number of individuals, we often have a Latin word that includes them all. Father, mother, brother, sister, son, daughter, husband, wife, child, are English, but fain ily is Latin. The distinctive name of every well-known living thing, or weed of field or forest, has an English name, but creature, animal, beast, plant, and herb are Latin. So we run, walk, leap, hop, creep, swim, ride, sail, turn, wheel and reel and totter and fall in plain Eng- lish ; but every motion is Latin. From a great variety of other sources we have acquired a few words, sometimes in very roundabout ways ; thus the name of the muscadine grape of the South is traced back through French, Italian, Latin, and Persian to Sanskrit. These are relatively trifles, but it remains to speak of another source of English, as important as any yet men- tioned, and that is Greek. As the New Testament was written in Greek, a few words, such as alms, baptize, cate- The Sources of English. 31 chism, Christ, heretic, hermit, have followed the Gospel wherever preached. But with these scanty exceptions, Greek was almost unknown throughout the Middle Ages, outside of the ever-shrinking Byzantine empire. Learning had forsaken its ancient seats, and when Pope Paul, in the eighth century, sent Pepin a present of what books could be found in Italy, the collection consisted of a Latin gram- mar, a hymn-book, and the forged works attributed to Dionysius the Areopagite. But while all real learning was banished from the West, the successors of Mahomet, under the guidance of the Jews and Nestorians, were eagerly acquiring it in the East. While a French priest was cir- culating a letter purporting to have been addressed to mankind by Jesus Christ, and brought down by the archangel Michael, Al Maimun at Bagdad was trans- lating Euclid, measuring a degree on the meridian, and determining the obliquity of the ecliptic. The Arabs found useful and congenial employment in collecting and translating the writings of the best ages of Greek literature ; and it was through the schools of Seville, Cordova, and Granada that these works found their first entrance into western Europe. From Spain the revived learning spread to the free cities of Italy, and from them was slowly dis- seminated through Europe. Greek was first introduced into the University of Oxford in 1500, eight years after the discovery of America ; and it was still forty years later when the first professorship was established at Cambridge. But neither those early teachers nor the ancient Athenians could have imagined the wide application to which the Greek language was destined in naming the objects and operations of modern thought. They could not have foreseen the searching analysis that was to be applied to every substance and movement in nature, to every tissue and function disclosed by organic life, every process and product of art, every operation and aberration of the human mind. He who finds a new thing has a right to give it a new name; and the consumption of Greek in giving names to things remote from the daily thoughts of men is enormous. The 32 The English Language. words of Greek origin, including all those belonging to special subjects, probably outnumber those from any other source, and in English exceed those from all other sources. The exact number can only be approximated ; but the following is such approximation : Descriptive of the animal kingdom 72,000 Vegetable kingdom 13,000 Sciences connected with medicine 18,000 All other subjects, perhaps 10,000 Total ii 3,000 The practice of forming scientific terms from Greek is, no doubt, in part a matter of habit and fashion, but, aside from these frivolous reasons, no other source would serve so well. It is not because Greek is a learned tongue, or that the Greeks had the words we now use. They neither had the words nor any use for them. We make them to order, just as we might make such a word as switch-tender-stand, which is an excellent example of the kind of words we should make of native material long, inelegant, and cumbersome. Greek has the advantage of combining with extraordinary facility into pronounceable compounds. Its consonants and vowels are not gathered into solid, insoluble lumps, but very evenly distributed, and upon a page are almost equal in number. This, I think, is the foundation of its excellence. The languages of northern Europe abound in undistributed con- sonants, strz, ntzsch, Idschm, krzyz. Hence, in combining several words into a new compound, each part is apt to begin and end with consonants, and the result is such a word as Grundungsschwindeln. Compared with such an unwieldy leviathan the longest term in Greek is a plaything. Skoro- dopandokeutriartopolis ripples along as pleasantly as a sum- mer brook on a pebbly bed ; and the farrago of Aristophanes, that contains 169 letters, moves so trippingly on the tongue that one might dance to it. We may illustrate this modern use of Greek by the familiar word geography, from ge, earth, and graph, write ; literally earth-writing. We see at once The Sources of English. 33 that the word we actually use is much neater than its Saxon equivalent. It is next observed that between the parts an o is inserted, that belongs to neither. The privilege of inserting at pleasure a connecting vowel facilitates greatly the making of new compounds. In this way we form just as easily geology, geodesy, geognosy, geogeny, geometry, geonomy, geo- phagy, and about 140 others, easily distinguished, easily pronounced, and to the learned of all nations disclosing their meaning at sight. It is even an advantage that their structure is seen only by the learned. They are thus kept to the form and signification intended, and are not corrupted, frittered away, and applied to whatever might happen. Dinotherium is literally terrible beast ; but the English equivalent could not possibly be kept as the name of a particular animal. There are patriotic persons who lament the loss of every Saxon word, and deem it matter of deep regret that our language ever admitted foreign elements. I do not, to any great extent, share their grief. With words as with men, present usefulness and good qualities far outweigh ancestral pedigree. Sugar is not less sweet, nor is its name harsher to tongue or ear, because it is a stranger from Arabia. We cannot, indeed, be quite sure how a language would sound that we never actually heard ; but, so far as I can judge from its appearance, Saxon seems to have been a cumbrous affair. While in no case more facile than modern English, it was often far more unwieldy. Take a few examples of the more unmanageable words : daeghwamlican daily leorning-cnihtas pupils, scholars msegen-thrymnesse glory modstatholnesse fortitude onbesceawian oversee unanbindendlicum inseparable I do not question but that words of any desired power might be constructed in this way, if only made sufficiently 3 34 The English Language. long and unpronounceable. Russian and German show what can be accomplished with native material. Here are a single Russian word and half a dozen German : Bolotnoperemezhdayushchagosya Erschiitterungssphare Geschwindigkeitsmesser halbkreisformiges Grubenschienenbahnwarter Kriegsverpflichtungsamt Verwandtschaftsnamen The German comes as near as possible to making his word a sentence and his sentence a metaphysical disquisition. The German language is unquestionably an instrument of great power ; but its power is a little like that of the bow of Ulysses, which was chiefly famed for the difficulty of using it, and was not half so effective as a Winchester rifle. Eng- lish has been saved from such productions as those just exhibited, and been made what it is by a thousand years of living contact with other peoples and tongues. If it be said that Saxon words might have been simplified by time, as many of their modern representatives really are, I answer that such a change was no doubt possible, but would it have taken place without the constant presence and pres- sure of foreign models? No such simplification has taken place in Russian or German. And further our words of Saxon origin are easy because they are short, and we have given up making them into long combinations. Looking then merely to the past and present qualities of the English language, I cannot but regard the Norman conquest as a great blessing. The English language has profited by its multifarious acquisitions because it has transformed them all after its own image. All but a few of the simplest words have been recast, and are no longer Saxon, Norse, Welsh, French, Latin, or Greek, but English. Every language has a character of its own a scale of sounds, an accentuation, a rhythm and The Sources of English. 35 cadence, a set of beginnings and endings, and a whole mechanism of speech peculiar to itself so that a good ear may distinguish one language from another without knowing a word of them. If our words had been preserved in their native forms, we should have an unspeakable piebald jargon instead of the harmonious unity seen in the Bible and the works of Lord Macaulay. It is well that our fathers followed the example of the diligent bees, that gather the juices of every flower and combine them into a homogeneous whole, sweet, nutritious, and wholesome. Our words are often so transformed that their own mothers would not know them ; and the average English speaker can no more distinguish them according to origin than he can the children of the Saxon and the Norman. Here are 60 words from 24 different sources, and none but the scholar would know that they are not all equally native-born : barrel chest frolic muslin skunk basket cider giant myrtle slag bishop clinker ginger paper spigot block cradle girl peach squirrel blue crease gum pepper swindle bonnet dirt hemp queer talk boy dish hurricane rose tea brindled dog husband rum trowel candle dog-cheap jacket sable vampire carpenter elm kale sabre viper cedar fellow lamp sack vow cheese fog measles silk whiskey At present the disposition is to maintain foreign words with all their foreign peculiarities unimpaired, as if it were more important to preserve their original nationality than to make them good citizens. The only reason of any perceptible force is the desirability of preserving the pedigrees of words ; but that is a matter that interests only the small number who have no need of such aid. We can scarcely avoid believing that the outlandish words are sprinkled in, and their unfamiliar sounds imitated to display the elegant 36 The English Language. acquirements of the writer or speaker. Far the greater number are French, and so have the least possible harmony with English. Why should such expressions as denouement, couvre-pied, coup d'ceil, be forced upon the English reader or hearer ? They are unnecessary, and jar upon a sensitive ear like discords in music. They are the stock-in-trade of inferior writers, and especially female novelists. The writer of " Robert Elsmere " might have taught all she had to teach and more, without introducing similar uncouth expressions 1 80 times in one small volume. Several estimates have been made of the relative number of modern English words derived from different sources. In the last century George Hickes estimated on the narrow basis of the Lord's Prayer, that nine tenths of our words were still Saxon. Widely different was the conclusion of Sharon Turner, that the Norman were to the Saxon as four to six. Dean Trench computes that 60 per cent, are Saxon, 30 per cent. Latin, including those received through a French chan- nel, 5 per cent. Greek, and 5 scattering. M. Thommerel, by counting every word in the dictionaries of Robertson and Webster, obtained the result that of a sum total of 43,566 words, 29,853 were of Greek or Latin origin, 13, 230 Teutonic, and 483 from all other sources. These discordant computa- tions seem to overlook the fact that many words that may be conveniently called hybrids cannot properly be charged to any particular source. They are of three classes. The first and largest consists of compound words like penman, peacock, pyroligneous, aldehyde, the parts of which are taken from dif- ferent languages ; the words of the second are derived from proper names, as Cartesian, Flemish, dahlia ; and a few are of uncertain origin. Skeat's Etymological Dictionary contains a little over 13,000 words, as it excludes generally obsolete and local words, derivative forms, and the technical terms of the more unfamiliar sciences. A classification of its contents, omitting a few duplicate forms, will give a very fair idea of the sources of the words employed in general literature and conversation. They are : The Sources of English. 37 Anglo-Saxon and English 2,863 Low German 1 16 Dutch of the Low Countries 187 Scandinavian 688 High German, of all periods 221 Teutonic, indeterminate 90 Celtic languages 351 Latin 2,094 Latin through French 3,545 Latin through other channels 341 French, not traceable farther 129 Proven9al (cJiarade) i Italian 43 Spanish 25 Portuguese (cocoa, dodo, emu, yam) 4 Greek, adopted directly and indirectly ^388 Slavonic languages 14 Lithuanian (talk] i Hungarian (hussar, sabre, shako, tokay] 4 Turkish 14 Persian 77 Sanskrit 39 Hebrew 72 Syriac 8 Chaldee (raca, talmud, targum) 3 Arabic 107 Other Asiatic languages 53 Oceanean 5 African languages 24 American languages 46 Hybrid words 419 Unknown . 21 Total 12,993 But a great part of these words are rarely met with, being either obsolete or confined to some special art or science. The vocabulary of almost any single author would show a much larger native element. The proportion of words from each of the principal 38 The English Language. sources varies greatly, according to the subject and the mental habit and associations of each speaker. Country folk talk in Saxon of their farms, crops, and families ; ladies of "society" prefer a liberal seasoning of real or supposed French ; literary people and those who aim at being sarcas- tic use a great many words derived from Latin ; and scientific specialists abound in Greek. Several methods have been proposed for analyzing the styles of different writers, but chiefly with a view to showing the extent of the native element, without further distin- guishing the others. One method has been to select a pas- sage, arrange the several words as in a dictionary, and count each one once. It is objected to this plan that a word that occurs only once, and so has little effect on the general style, counts for as much as one repeated a hundred times. It has therefore been proposed as a second method to count each word as it stands on the page, making " the " perhaps equal 30, and " flocculent," I. The results of these two meth- ods would be very different. But the words of most fre- quent occurrence are little ones that by themselves suggest no idea ; and these are always native English and are always present, whatever the style may be. They are a constant element, and, if counted, disguise the actual differences of style. There is therefore a third resource, to count only the significant words those that represent things, actions, and qualities, including those secondary qualities that de- scribe actions and other qualities, as rapidly, slightly. To make this clearer let us take a sentence and italicize the significant words : " Many cabals were formed, loud complaints were uttered, and desperate resolutions taken ; but before they proceeded to extremi- ties they appointed some of their number to examine the powers in consequence of which the cardinal exercised acts of such high authority" ROBERTSON : " Charles V." Here are 40 words, of which just one half are insignifi- cant and such as have to be used by every one. Of the The Sources of English. 39 other 20 cabal is Hebrew; loud, high, uttered, and taken are native English, and the remainder derived from Latin, either directly or through a French medium. According to the second method of computation the native words are 60 per cent. ; by the third they are 20. The last method is the one adopted in the following table, which is calcu- lated on a basis of 200 significant words divided into two classes, native and foreign. Proper names are passed by altogether. The numbers give the percentage of native words. Kentish Sermon A.D., 1250, 92 ' Havelock the Dane " 1300, 87 Sir John Mandeville ' J356, 69 Chaucer, Prologue, Cant. Tales " 1390, 58 Wycliffe (Luke xx.) " 1389, 70 Tyndale (Luke xx.) " 1526, 63 Authorized Version (Luke xx.) " 1610, 64 Ghost's Story in Hamlet " 1600, 49 Bacon, Essay 29 " 1612, 30 Dryden (prose) " 1683, 29 Dr. Johnson " 1748, 24 Gibbon " 1776, 18 Hawthorne " 1853, 43 Macaulay (History) " 1848, 33 Dasent's Translation of the Gisli Saga. . " 1846, 81 The native English element underwent a pretty steady decline from the age of Edward the Confessor to the close of the last century, reaching the lowest point in the ponder- ous sentences of Gibbon, in some of which every significant word is Latin : " From such laudable arts did the valor of the Imperial troops receive a degree of firmness and docility unattainable by the impetu- ous and irregular passions of barbarians" In the present century there is a reaction against this mlemn, labored style and in favor of a freer use of native English. The result has been a style more crisp, fresh, and 40 The English Language. direct. The writer who, so far as I know, has carried this use of English furthest is Dr. Dasent, distinguished by his illustrations of Icelandic literature. As a contrast to the style of Gibbon I present a passage from his translation of Njal's Saga, introducing a famous chief and warrior Gunnar of Lithend : " He was a tall man in growth, and a strong man best skilled in arms of all men. He could cut or thrust or shoot if he chose as well with his left as with his right hand, and he smote so swiftly with his sword, that three seemed to flash through the air at once. He was the best shot with the bow of all men, and never missed his mark. He could leap more than his own height with all his war-gear, and as far backwards as forwards. He could swim like a seal, and there was no game in which it was any good for any one to strive with him ; and so it has been said that no man was his match. He was handsome of feature, and fair skinned. His nose was straight, and a little turned up at the end. He was blue-eyed and bright-eyed, and ruddy-cheeked. His hair thick, and of good hue, and hanging down in comely curls. The most courteous of men was he, of sturdy frame and strong will, bountiful and gentle^ a fast friend, but hard to please when making them." The simpler the ideas are, and the nearer to every-day life, the greater the share of English words that may be used. Sentences made up wholly of native words are common enough in conversation, but are rare in books, except in the Bible. See as examples Job xxxi., 21, 22 ; John i., 1-4. Sentences containing no native English words are still rarer. Here is an attempt at constructing one : " Injudiciously profuse eleemosynary aid, defeating benevolent intentions, frequently stimulates voluntary pauperism." The question how large a part of their native tongue most people know is answered by Professor Max Miiller in his usual clear and decisive manner : " A well educated person in England, who has been at a public school and at the university, who reads his Shakespeare, the The Sources of English. 41 Times, and all the books of Mudie's Library, seldom uses more than about 3,000 or 4,000 words in actual conversation. Accu- rate thinkers and close reasoners, who avoid vague and general expressions, and wait till they find the word that exactly fits their meaning, employ a larger stock ; and eloquent speakers may rise to a command of 10,000. The Hebrew Testament says all it has to say with 5,642 words ; Milton's works are built up with 8,000 ; and Shakespeare, who probably displayed a greater variety of expression than any writer in any language, produced all his plays with about 15,000 words." " Lectures on Language," lee. vii. This is not precisely saying how many words we know, but as eloquent speakers may only rise to the command of 10,000, it may be fairly presumed that common folk could not use so many if they would. The qualifications of his lowest class are so high, that most cultured Americans would certainly be limited to not more than 4,000 words. Very few of us have had the advantages enumerated. But approaching the ques- tion from another side, we find Webster's Dictionary edition of 1890 professing to contain over 1 1 8,000 words. Joining these two points, we should reach the remarkable conclusion that a moderately educated American is ac- quainted with less than -fa of his mother tongue, which is manifestly and widely erroneous. The term word is not used in the same sense by both authors. The Oxford pro- fessor is doubtless a better guide in this than the American lexicographer. The enormous number of words stated in the Dictionary gives a false impression. Some of these words are completely obsolete, others are still entirely foreign expressions, others again are restricted to particular arts or sciences, and form no part of the language of ordinary life or literature. Examples of these words which might prop- erly be omitted are : qa ira, bedagat, cackiri, caimacan, cous- cous, Davyt, dawm, dendrocolaptes, doand, couzeranite, Ich dien, ad quod damnum, (Zgrotat,pulmonibranchiate, angiomono- spermous. Many words are merely various spellings of the same. Many are repeated as nouns, adjectives, verbs transi- tive and verbs intransitive. Winter and water are each 42 The English Language. counted as three words. The obsolete dogly is made to do duty as two words, and the equally antiquated dorr as seven. Words are drawn out in long array by means of suffixes and combinations. The simple word delight counts for 16 ; 25 combinations are made with honey ; and water is an element in 203. Some of these combinations are such as well-water, well-borer, winter-apple, winter-wheat, which every one can make for himself ad libitum and ad infinitum. Most of these words may be very properly placed in the Dictionary, but their reiteration gives an undue air of immensity to the language. On 40 pages taken at random I find 2,741 words, of which 357 may be reckoned as obsolete, foreign, or tech- nical expressions. Of the remainder 1,635 are repetitions, derivatives, or compounds, that retain the primitive significa- tion, leaving 749 really distinct words. Then taking my own knowledge, as the only measure of the general intelligence that can be applied, I find 209 with which I am not ac- quainted, which is less than one third () of the primary words, or one eleventh (yV)> including the derivatives and compounds. The outcome of all which is that instead of knowing but an insignificant fraction of our language, we are more or less familiar with over five sixths ($) of the words and forms of words available for general literary purposes and daily use ; also that with the above limitations Webster's great Dictionary probably does not contain more than 30,000 independent words. CHAPTER III. THE PROVINCE OF GRAMMAR. ENGLISH grammar has long been defined as the art of speaking and writing the English language correctly. One of the text-books used in the public schools of this city calls it more ambitiously, " the science which teaches how to speak and write the English language correctly." The claim is a large one, and not to be conceded without inquiry. " Our language," says Professor Whitney, " like every other, is made up of words." Of these, Webster's Dictionary, edi- tion of 1890, professes to contain "an aggregate of upward of 118,000." One who knows all these thoroughly, or even 20,000 of the most necessary, and how to use them, possesses the art of speaking and writing the English language cor- rectly. The art should certainly include a knowledge of the words themselves. Let us see : we may select almost any word and inquire what the grammatical text-books have to tell us about it. We may take the word theatfe, and inquire its origin and history, its meaning, its form theatre or the- ater, its pronunciation theatre or thedtre. As a mere word, this is nearly all that we care to know about it. To all these questions, a good dictionary will furnish answers; a grammar will give none. From the grammatical text-books we may infer one point as probable that when we speak of more than one such establishment we should add an s. But that simple fact is known about as well by the street Arab as by the graduate of the highest schools. In justice, however, it must be conceded that when such a distinction is uniform, or nearly so, a considerable labor of search is saved by assuming it to hold good, and taking the chances. 43 44 The English Language. Unfortunately the assumption would fail us just where it is most needed in unfamiliar words, as anas, amaryllis, incubus, polyergus. No grammar will help us to distinguish the lum- bar region from the lumber region, or discriminate between the expressions, to differ from and to differ with, so that in nearly all cases of difficulty we must have recourse to the dictionary and not to the grammar. . If lion were the name of the male of a certain species of animal, and lioness were the female, and the same held good universally, it would be quite convenient, as on learning one form of the word we could readily infer the other. But this relation is of so rare occurrence, that a knowledge of it is of no practical value. Again, in Arabic for example, in words of more than one syllable the last never has the accent ; if the next to the last ends in a consonant or written long vowel, it is accented ; otherwise, the second from the last. Here the whole system of accentuation for the language may be expressed in a single sentence. Icelandic is still simpler in that respect, for there the first syllable is always accented, be the word long or short. But no such absolute rules obtain in English, where the accent of each word must be learned by itself. In short, ours is a language of exceptions and irregularities, in which the dictionary counts for every- thing, the grammar, almost nothing. But if, from the irreg- ularities of our language, grammatical rules are of narrower scope than in some others, we shall find that we have less use for them. It has been hinted above that plural names names of more than one have generally an s added to dis- tinguish them. There might well be other additions or changes corresponding to modifications in the meaning of words. In point of fact there are a few such in English. The orderly presentation of these changes in names gram- marians call declension ; in words expressing action of any kind, it is called conjugation ; and these comprise the larger part of grammar. Professor Max Miiller says rather abso- lutely, " What is grammar after all but declension and con- jugation ?" It is unquestionably true that they constitute the greatest part of all that is of immediate practical value. The Province of Grammar. 45 But some languages may be richer in varying forms and require more declension and conjugation than others. In English a name, or noun, as it is called, may assume four forms, thus : man men man's men's None present a fuller declension than this, and few are so complete. If boy had been selected instead of man, while by orthographic expedients four forms might have been pre- sented to the eye boy, boys, boys, and boys', to the ear there would have been but two. So in other languages some of what are theoretically different forms are no longer dis- tinguishable ; but, counting the full number, a Latin noun has 12 variations, a Greek 15, a Hebrew 26, while a Hunga- rian or Magyar, with its various affixes, admits of 1,154 com- binations. Or, if we take such a qualifying word as earthen, or English, which with us has only one form, its synonym in Latin might have 36, and in Greek 45 variations. Again an English verb never has more than 8 distinct forms, and sel- dom more than half that number in actual use. Of the entire system, as write, writest, writeth, writes, writing, wrote, wrotest, written, three are practically obsolete ; while far the greater number have in use only four forms, as sail, sails, sailing, sailed. In contrast with this scanty stock, the Spanish verb presents (theoretically) 120 variants ; the Latin 444; the Greek according to Kuehner's Grammar, 1,138, ac- cording to Professor Miiller the round sum of 1,300; the Hebrew 246; and the Arabic over 2,100; while Professor Whitney cites the Rev. T. Hurlbut as saying that he had ascertained by actual computation that an Algonkin verb admits of 17,000,000 variations. If then grammar be merely declension and conjugation, which is not far from the truth, it plays comparatively a very insignificant part in English. All the irregularities of our language are more than compen- sated by the extreme paucity of its grammatical forms. It is almost as grammarless as Chinese, in which no written word is ever varied by a single stroke or dot, and when 46 The English Language. spoken admits of only a change of tone. The weary hours and years spent by our youth in parsing English sentences according to forms borrowed from Greek and Latin are worse than wasted useless for the avowed purpose of learn- ing to speak and write, and leading to a misapprehension of what our language is. Professor Whitney, in his " Essen- tials of English Grammar " says, " Nor is the study of the grammar of one's native tongue by any means necessary in order to acquire correctness of speech. Most persons learn good English in the same way that they learn English at all namely, by hearing and reading." The same opinion is probably held now by all competent persons who have given the subject attention. We know, too, that many of the masterpieces of human literature, in languages incomparably more intricate than ours, are the work of men who had never heard of grammar as either a science or an art. All that is most prized in Greek literature was written before any book on grammar had been seen west of the Euphrates. Only two peoples of all the world, the Hindoos and the Greeks, originated the idea of analyzing their languages and codifying their peculiarities. The former had the priority in time, and the superiority in analytical acuteness. To learn to write and speak correctly was not the object of either. They could do that already ; but both wished to preserve unaltered their oldest and most revered writings. The Brahmin re- garded the hymns of the Veda as no mere human composi- tions, but only seen in vision and copied by the ancient Rishis. To him the correct pronunciation and accent of a syllable might determine his salvation. Hence immense labor was spent in observing and noting the form of every word and the recurrence of every change. In the fourth or fifth century before Christ, and long before the invasion of Alexander, the Hindoos had traced all the words of the Sanskrit to 1,706 roots, and determined the particles and affixes with which they were combined, and all the outlines of grammar as now understood. The thought of grammar had not yet occurred to the Greek. Plato, in his philosoph- ical speculations and with no view to correctness of style, The Province of Grammar. 47 divided words into nouns, onomata, and verbs, rhemata ; although we are left in the dark as to what each class con- tained. He was also acquainted with the distinction between vowels and consonants. Aristotle, for merely rhetorical pur- poses, added the classes of conjunctions and articles, but by the latter he meant pronouns and relatives. This is as far as Greek grammar advanced in its native country. But before Hindoo or Greek had made the analysis of his language the nations of Chaldea had compiled dictionaries and grammars for the more practical purpose of learning a language not their own. The great kingdoms of which we read in the Bible had been preceded by a cultivated people of wholly different speech a people devoted to science and literature, who had left behind them considerable writings, or rather printings. The speech of that early people was to those who came after what Latin was to Europe for so many ages, and text-books were prepared for learning it. Modern explorers find fragments of those ancient grammars among the ruins of the royal libraries of Sargon and Assur-bani-pal. 1 Early civilization had three chief, and perhaps independent, centres of development, China, Chaldea, and Egypt. The first two had but little influence on the rest of the world ; but Egypt, lying on the Mediterranean, the great highway of nations, was the mother of western science, whose cradle was rocked by the Pharaohs and the Ptolemies. In the new city of Alexandria men of all climes, from India to Spain, and from Mount Atlas to Norway, met and exchanged the products and ideas of their countries. The halls of the great Museum were, thronged with more than 10,000 students. The director, Demetrius tPhalereus, had orders to collect all the writings of the world for its libraries. The copies of the old Greek classics were found to 1 Sargon I. looms in the dim distance like the figures of Haroun Al-Raschid, Charlemagne, and Alfred. Assur-bani-pal is celebrated by Arrian and Nicolaus of Damascus as ' ' Sardanapalus, King, and son of Anacyndaraxes, who built two cities in one day, Anchialus and Tarsus." He was really the son of Essar-had- don and grandson of Sennacherib II. (Kings xix., 37), the prototype of Louis XIV., and among other things a munificent patron of learning. 48 The English Language. contain various readings. Out of these the scholars of the Museum undertook to publish critical editions. This neces- sitated a minute study of the text. Zenodotus, the first librarian, about B.C. 250, detected and pointed out the per- sonal pronouns and the singular, dual, and plural numbers. Nearly a century later Aristarchus discovered the preposi- tions. In all this there was no thought of developing an art to teach people how to use their mother-tongue. But when it became the fashion for the young Roman gentry to learn Greek, as the moderns do French, a necessity arose for systematic analysis. How could the 1,138 parts of the Greek verb be reconciled with the meagre ^/j/] of the Latin, unless they were tabulated and explained. This work was reserved for Dionysius Thrax, a pupil of Aristarchus, who came to Rome as a professor of Greek in the century before the Christian era. For convenience he reduced the sub- stance of his lectures to book form the earliest European treatise on grammar, and still extant. From this work the distinctions and terminology of the Greek language were afterwards, by translation and mistranslation, applied to the Latin. The Latin language was stretched upon the iron bedstead of the Greek, as the languages of modern Europe have long been racked on that of the Latin. The ideas of grammar were not applied to English until about the time of the Tudors, and then under the impression that every human dialect could be laced in the harness of the Latin. The first text-books used in the schools were written in that language, and designed, not so much to teach English as grammar in the abstract, as applicable alike to all. They of course taught only Latin. The most famous of these early grammars was that of William Lily, first printed in 1542 by express authority of Henry VIII., and long known as " King Henry's Grammar." Although called simply " A Grammar," and not the grammar of any particular language, it was not only written in, but related exclusively to, Latin. And when the same author subse- quently published a dense black-letter volume with the deceptive title of " Lilie's English Grammar," it was but a The Province of Grammar. 49 collection of the rules of the Latin strung on a thin thread of English text. The first really English grammar was claimed by William Bullokar, who published in 1586 "A Bref Grammar for English/' which he said was " the first Grammar for English that ever waz except my Grammar at large." Since that time there have been English grammars innumerable, not a few of them written in Latin, as those of John Wallis and Charles Cooper, in the time of William and Mary, and with rare exceptions ignoring the real source and character of the grammar of the language. Even so late as 1796 the greatly improved grammar of Thomas Coar, published in London, filled its pages with diagrams like the following: SINGULAR PLURAL NOM. a house NOM. houses GEN. of a house GEN. of houses DAT. to a house DAT. to houses Ace. a house Ace. houses Voc. O house Voc. O houses ABL. with a house ABL. with houses It would have been quite as easy and rational to have added a dozen more prepositions, or to have omitted the most of these, but that this was the scale recognized in Latin. Eminent men, not the authors of systematic treatises on grammar, have sometimes furnished valuable suggestions. Roger Bacon pointed out the folly of trying to explain words by reference to some remote language with which they had no connection. Locke expressed his conviction that all words, if they could be traced to their sources, would be found to have originally denoted visible objects,, their sensible qualities and actions. Following up this hint, John Home Tooke published in 1786 his famous " Epea Pteroenta," or " The Diversions of Purley," a work of singu- lar acuteness and ingenuity, designed to show that all our little words of scarcely perceptible signification if, and, but, although, etc. are the relics of once substantial nouns and verbs, and that their source is to be sought, not in 4 50 The English Language. Greek and Latin, but in the earlier forms of English and the closely allied languages. Although from rashness and imperfect knowledge, Home Tooke was often wrong in his derivations, and would be a very unsafe guide, still he pointed the way for all subsequent investigators. The first grammarians, as Dionysius, undertook to teach the signification, the spelling, and the pronunciation of words; but that has long since been turned over to lexi- cographers. This leads to consider again the distinction between the two classes of books. The dictionary treats of single words, and one at a time. When it has told us all it has to say of any word, as almanac, boom, yacht, we are not thereby helped to understand the next word that occurs. The grammar, on the other hand, deals with classes of facts or of words, and points out their distinctive agreements and differences. In the sentence " The swallow flies about catch- ing flies" the word flies occurs twice, presenting no differ- ence to the eye or the ear. The grammarian sees a difference in the meaning or application, and should at least try to discover whence came the s that is common to both. By vary- ing the above illustration we might say, " Two swallows fly in pursuit of one^fj/." The curious question would then be presented why the form that is singular for nouns is plural for verbs, and the reverse. Whatever result the grammarian might reach in this case would be equally appli- cable in thousands of others, and would be recognized as a general principle or rule. If the ed in march^ refers to the past, so it does in other words, and if its history and signifi- cance be discovered in one case, the discovery is equally good for all. Grammar then treats of everything relating to a language that can be reduced to general facts, principles, or rules. It has to deal chiefly with the various forms assumed by the same words. This is, in English, a very narrow field, but extremely rocky. Grammar, like botany or mineralogy, is a purely descrip- tive science. The duty of the grammarian is not to invent or create, but to state and classify the facts as he finds them. What is true of nothing else is true of language, that whatever The Province of Grammar. 5 1 is is right. Expressions may be intricate, awkward, incon- sistent with other expressions, difficult for the tongue or harsh to the ear, but so long as they are the unmistakable symbols of certain ideas, they answer their purpose. But there may be different and conflicting expressions for the same idea. One class of speakers say them is, while another say those are. The grammarian may indeed point out that the latter phrase is the most consistent with general usage, and is employed by the most careful speakers, and there- fore preferable ; but if the former were the sole recognized form, we should have to put up with it ; and it would be as absurd to object that it was ungrammatical as to accuse some wild-wood flower of being unbotanical. But it may be objected, " If grammar does not make rules for the government of language, and people can learn to speak and read without it, what is its use?" I readily admit that these considerations deprive it of a fictitious im- portance long attached to it, but it still retains a real value rarely thought of. That great body of knowledge known as learning is valuable indirectly rather than directly. By it are formed habits of calm, thoughtful observation and dis- crimination that modify the whole character of man. If a savage could be induced to give his attention for half an hour to the drawing of a circle, to the equality of its radii, its relation to the hexagon, the ratio of its inscribed and circumscribed squares, he would be a little less of a savage all his life after. All honest pursuit of knowledge has this humanizing effect. The world has more faith to-day in its men of science than in its princes, prelates, and states- men ; and yet a great part of science has no practical appli- cation. It requires but a small part of astronomy to find a ship's place at sea, or locate a boundary, and that small part is about all that touches his material interests. Minute examinations of protoplasm, of annelids and bacteria, are commended as science, even by men who would scout phi- lology as a waste of time upon " mere words." Yet if man is the highest of earthly creatures, and language his most dis- tinctive attribute, that too may merit some attention. Sir 52 The English Language. Samuel Baker thought he could distintinguish nine distinct caJls a language in embryo used by the baboons of Abyssinia ; and Dr. Charles A. Abbott claims to have found twenty-seven separate caws among the American crows. 1 The verification of these cries would be esteemed as a valuable contribution to natural history, even if we should never have occasion to converse with crows or baboons. A much higher interest attaches to the study of language. Like everything else in these days, it is the result of growth and development under conditions and laws that can be in part ascertained. Each word has a pedigree reaching back to the times of the paleolithic cave-dwellers. The philolo- gist may be compared to the geologist found poring over a gravel bank or a ridge of disjointed stones, who explains to the curious wayfarer that each pebble or block has a history of its own a part of the history of our planet. Not one of them is a native of this place, or has the form it once bore. They have been torn from their distant beds by successive convulsions or slow upheavals, rolled for ages in currents of water until their angles are worn off, borne across seas by drifting ice, or dragged snail-pace over the land by glaciers. Here, side by side, lie fragments of granite and quartz, with silurian slates and limestones, breccias, porphyry, and basalt. We can trace the track of some ; and were our knowledge complete we could find the distant source of each. From the mould overlying some, their coming must have been long ago. Somewhat similar is the position of language. It has been intimated already that the chief part of gram- mar is declension and conjugation, and that these in English are scanty. Indeed, our language offers but the scattered remains of an inflexional system. In this respect the inflec- tional systems of Latin and Greek, for example, might be likened to skeletons set up in a museum, bloodless and life- less, indeed, but with every bone and joint in its place. On the other hand, English presents only here and there a bone, so broken and worn as to be identified with difficulty and only by comparison with the appropriate skeleton. 1 Popular Science Monthly, vol. xxv., p. 475. CHAPTER IV. WORD MAKING. THE reading public, at least of this country, owe much to Professor Max Miiller. He has furnished them a great amount of philological information in a very attractive form, and he has given every one something that he can take exception to. Among others is the dictum that " No man ever invents an entirely new word." It is true, indeed, that the scholar, with a wealth of words, ancient and modern, will rarely contrive a new combination of sounds, which would necessarily be unintelligible. But pure invention comes of poverty rather than of riches, and those who have fewest words have the greatest temptation to invent. Little chil- dren often form new combinations, that become for a time household words, while a few obtain a wider circulation. The very illiterate are prone to do the same, and it is prob- able that many low words, like bamboozle, cavort, doggerel, splurge, scalliwag, and mulligrubs, originated in this way. Mormon, now familiar to the whole world, is an entirely fac- titious word. Persons with no lack of words occasionally amuse themselves by contriving new ones. Opodeldoc is an invention of Paracelsus, and Decandolle, the Swiss botanist, devised sepal to designate a division of the calix of a flower. The word quiz was introduced by the keeper of a Dublin theatre named Daly, on a bet that a new word of no mean- ing would become the town-talk in twenty-four hours. He gained the wager by setting boys to chalk the word upon walls. Van Helmont proposed two new words gas, to de- note a form of matter then attracting attention, and bias for a supposed influence of the stars. The world having use for 53 54 The English Language. the word gas, and not for its fellow, the former has had uni- versal acceptance, the latter total neglect. Darwin or Huxley could not find a clearer case of survival of the fittest. Still, with few exceptions, all the words we are likely to meet with are made up of modifications or combinations of previous words. Doubtless there was a time when all the words of our early ancestors were few and simple, each consisting of a single vowel a, i, o a vowel preceded by a single consonant do, be, go, no, say or a combination of the two, as ado, ego, era, In many languages the words are still chiefly made up of alternate consonants and vowels. In all the Polynesian tongues no syllable ends with a consonant, and two conso- nants never come together. The words are formed on the pattern of ta-bu, Ta-hi-ti, la-ve-na, Ho-no-lu-lu. In many of the native American languages this habit is quite observable. In 50 pages of the Hidatsa Dictionary of Dr. Matthews I find but 30 syllables that end with consonants. Many other languages show a dislike to certain consonantal endings. Every Chinese word is a single syllable, ending either with a vowel or a nasal. Greek admitted no final consonants except n, r, and s ; the French suppresses many in pronun- ciation ; and modern German discards the sounds of final b, d, g, v, and z. It is probable that wherever several conso- nants come together intervening vowels have been suppressed. In respect to signification, these earliest words were prob- ably nouns, adjectives, or verbs that is, they represented things, qualities, or actions indifferently as occasion might require. If we say that a man brings a saw&nd a jaw-horse to saw a load of wood, we employ the same word succes- sively as noun, adjective, and verb. In English, this free- and-easy way of playing at rights and lefts is very common and very convenient, as when we say a bean-pole and a pole- bean, a cart-horse and a horse-cart. In Chinese, which is the most primitive form of human speech, and therefore one of the most instructive, every word is of one syllable, and every syllable a word, unchangeable except in the tone of utter- ance ; and grammatical distinction is unknown. Thus ta, as Word Making. 55 a noun, is greatness ; as an adjective, great ; as a verb, to grow or be great ; and as an adverb, greatly. Of these primary Chinese monosyllables there are reckoned 450, and it is thought that no language has a much larger number of ultimate elements. These primitive syllables are often called roots, as many of them have produced abundant crops of derived words. We may get an idea of a root sufficient for our present pur- pose by taking such a series as corn-pel, d\s-pel, ex.-pel, im- pel, pro-pel, re-pel, and calling pel a root with the sense of drive. The other syllables in this instance are termed pre- fixes ; if they followed the root, they would be called suffixes. The root itself may undergo changes, as in com-//-sory, im- pul-s\vQ. As the same root may run through a large family of languages, it can sometime^ be traced in several thou- sand combinations. The same prefixes and suffixes may be attached to many different roots : thus zw-tend, z'#-spire, in- fliction. While the earliest words, no doubt, represented material visible things, their qualities and actions, one may see, by turning over the leaves of a dictionary, how they become freighted with secondary and figurative meanings. In this way all terms for abstract and immaterial things are ob- tained. The original meaning is often lost sight of, and the derived alone remains. Thus spirit once meant only breath in-spire and ex-pire, to breathe in and breathe out compre- hend was to grasp and hold together, and disgust was a bad taste in the mouth. At a very early period, too, a few roots must have assumed a pronominal character. They became hieroglyphic or short-hand expressions, denoting in them- selves neither things, qualities, nor actions. They were in signification such words as here, there, this, and that, accom- panied no doubt with the act of pointing with the finger. But while we should naturally expect our first ancestors to have busied themselves like Adam in giving " names to all cattle," it is remarkable to observe that the tendency of research in the most developed languages the Aryan and Shemitic is to show that the oldest words that can be 56 The English Language. reached represented, not things, but qualities and actions. Still we must remember on the other hand that the vedic hymns and the martial ode of Debprah are relatively little nearer the beginning of things than we are. I have said that Chinese retains the most primitive char- acter, consisting of single syllables, every one uttered sepa- rately as a child begins to read. Most other peoples think this too slow. Not content with learning to do a thing, they want to do it quickly and with little labor. Let us take a French sentence as an example " Tu as ce que il te faut" No Frenchman utters this as seven separate words. What he does say is something like, " Tua skeelt fo," and his whole language is similarly compressed. The motive is to so great an extent the saving of labor that laziness has been recog- nized as one of the chief factors in the production and con- fusion of tongues. Words that happen to be used often together come to be combined and pronounced as one. This is a gradual process in which we easily distinguish three steps. In brick house we have two distinct words, but brick has become an adjective descriptive of house ; in work-house two words are treated as one, but to show that they are not yet perfectly consolidated, a hyphen (-) is placed between them. The first part is uttered forcibly, the second lightly. The greater stress is called accent, and the two parts have but one that is, they are accented as one word. When we come to householder we are no longer notified that the parts were ever separate. Under which of the three forms we shall find any combination depends on length and frequency of use. Turnspit is written without, and turn-table with, a hyphen, because the English people have been much longer used to roasting meat on a spit than to turning railroad cars on a table. But laziness will not rest here. Compound words must next be shortened. The Danes have a pair that are very handy faster, a father's sister, and moster, a mother's sister. Comparable to these are the gaffer and gammer, for grand- father and grandmother, that used to be commonly heard in the West of England. To take a few more miscellaneous Word Making. 57 examples, the Portuguese coin moidore is moeda de ouro ; priest is reduced from presbuteros, bishop from episcopos, and alms from eleemosune. Our simple word which once had an / in it, and the Saxons commonly called it hwilc, which was itself an abbreviation, the fuller form being hwilic, what-like. If we go back to the fourth century we shall find that the Goths had a still fuller form, hweleiks. Our fathers in the days of the great Alfred, in praying for their " daily bread," took time to call it daeghwamlican hldf. Let us see what we have made of these two words. We will take the last first. We have thrown away the initial h, turned the Saxon long a according to our wont into a long 0, and now write the word loaf. From the more formidable looking word we have dropped the termination an, the g of daeg, a day, and hwam, meaning each. We have also reduced lie, like, to the now unmeaning ly. Taking these successive amputations in the order named, we should see the word as daeghwamlican, daeghwamlic, daehzvamlic, daelic, daely. These changes were not all made in a day. The tendency here illustrated is not exceptional. All languages that are at all developed are full of it. In fact that is what development means. The effect is sometimes curious. We will take as an example the French aujourd'hui, meaning now, to-day. First separate it into aujour d' hui ; next observe that au = a le = Latin ad ilium, that jour is from the Latin diurnus, that d' = de, and hui is Latin hodie = hoc die. Treated in this manner it can be stretched out into the very low Latin of ad ilium diurnum de hoc die. In like manner m$me is discovered to be a desiccated preparation of semetipsissimus. But such choice specimens are not confined to French. Old authors give an English phrase of the seventeenth century which they write muskiditti, meaning much good may it do you ; and Shakespeare has godigode'n for God give you a good even. The modern editions naturally give it quite incorrectly ; see " Romeo and Juliet," Act iii., Scene 2. But the masterpiece of all is the one most common. The final m of the house- maid's hourly Yesm is all that remains of the once dignified me a domina. 58 The English Language. Languages differ greatly in their aptitude for forming compound words. Chinese does not admit of them at all ; Spanish has few of native growth ; French has less ability to form them than English ; and this last has to a great extent lost the habit. There are three European languages Ger- man, Russian, and Greek that have almost unlimited capa- bility of forming new verbal combinations. Owing to the excess of consonants, German words are apt to be unwieldy, like Einwanderungsgesellschaft and Unabhdngichkeitserkla- rungen. In most instances each section begins and ends with consonants, and, in the language of working mechanics, the joints show. English labors under the same disadvantage, but to a still greater extent. Milk-maid may as well be deemed two words as one. It is not an indivisible whole, like the Greek derivations astronomy and geology. It is partly from this cause, and partly from the early acquired habit of adopting French, Latin, and Greek terms, that for the higher purposes of literature and science we rarely form a new word from native material. English, as we know it, is doubtless a very noble language ; but well it may be, for it has at command all the resources of at least three. Confined to the original Saxon, it would be very far from what it is. By long use and attrition, words quite diverse in their origin come to be written or pronounced alike. These homonyms, as they are called, are quite numerous. The fol- lowing, although less than a fiftieth part of them, will be sufficient to make their character intelligible : bay, i. Old French fai, Lat. badius, reddish brown. 2. Fr. bate, Lat. bacca^ a berry a kind of laurel tree. 3. Fr. bate, Lat. baia, an inlet of the sea. 4. Fr. abboyer, to bark. cleave, i. Anglo-Saxon cleofan, to split. 2. A.-S. clifian, to adhere. dock, i. Norse dockr, the tail to cut short. 2. A.-S. docce, a plant. 3. Old Dutch dokka, a place to lay ships. fell, i. The past tense of fall. 2. A.-S. fellan, to cut or knock down. Word Making. 59 3. A.-S. fel, fierce, destructive. 4. A.-S. fell, the skin. 5. Norse fjall, a mountain ridge. gill-, i. Norse gjolnar, the breathing organs in fishes. 2. Norse gil, a ravine. 3. Old Fr. gille, the fourth part of a pint. 4. Lat. Julia, also the ground ivy, nepeta glechoma. let-, i. A.-S. latan, to permit. 2. A.-S. lettan, to hinder. In languages having short words homonyms abound, and sometimes greatly embarrass the learner. In monosyllabic Chinese they are his principal difficulty. In words like wind-mill and horse-back the elements remain as complete and distinguishable as in Chinese ; but in such a word as un-kind-ness there are two parts that no longer exist as separate expressions. Most persons would surmise that the prefix un implied negation ; but many would be puzzled to assign a meaning to the suffix ness. Language is full of prefixes, suffixes, and interpolated syllables and letters that have no longer any independent life of their own, but cling like parasites to the more obviously significant parts of words. In un-sym-metr-ic-al-ly the main root of the word is metr, to which are attached two prefixes and three suffixes. The origin and meaning of some of these affixes can be traced, of some conjectured, and of others not even guessed. The following are the principal prefixes that occur in English : a-, i. Greek , without acephalous, amorphous. 2. Lat. a, shortened from ab, from, by, with amanuensis, avert. 3. Lat. ad, to ameliorate, astringent. 4. Lat. e for ex, from amend from emendare, through Old Fr. amender. 5. Gothic us, ur, Norse or, forth arise, awake. 6. A.-S. of, from adown ; A.-S. of dune, from the hill. 7. A.-S. and, over against, like Gr. avrl along. 8. A.-S. on, on, in, at afoot, aground, asleep. 9. A.-S. dn, one apace, apiece. 60 The English Language. 10. K.-S.ge, without any appreciable signification aware ; through gewaer, ywar oryewer, iwar, aware, 11. Norse at, to ado. 12. Fr. a, to achieve, from a chief, Lat. ad caput. 13. Fr. he, interjectional alas ; Fr. httas. 14. Dutch houd, hold avast, from houd vast, hold fast. 15. Dutch aan, to, towards aloof. 16. Arabic al, the apricot, introduced by the Portuguese. ab-, i. Lat. ab, from abjure, aberration. 2. Lat. ad, to abbreviate. abs-, Lat. abs, from abscond, abstract. ao, Lat. ad, to access, accommodate. ad-, i. Lat. ad admire, administer. 2. Lat. ab advance. The " d " is an interpolation of about the year 1500. The word was previously written avance, Fr. avancer, from Lat. ab ante. adv-, Lat. ab advantage. af-, Lat. ad affix, affidavit, affront. ag-, Lat. ad aggregate, aggravate. al-, i. Lat. ad alliteration, alluvium. 2. A.-S. eal, all alone, altogether, always. 3. Span, el, the alligator, i. e., el Jagarto. 4. Arab, al, the alcohol, algebra, alkali. am-, i. Lat. ad ammunition. 2. Lat. in ambush. 3. Lat. am, shortened from ambi, around amputate, to prune around. amb-, Lat. ambi ambient, ambition. ambi-, i. Lat. ambi ambiguous. 2. Lat. ambo, both ambidextrous. amphi-, Gr. aftfpi, around, on both sides amphitheatre. an-, i. Gr. av, Eng. un anarchy, anhydrous, anodyne. 2. Lat. ad annex, annul. ana-, Gr. ava, up, back again, reverse anatomy, anagram. ant-, Gr. avri, against antacid, antagonist. ante Lat. ante, before antedate, antediluvian. anti-, Gr. avri antidote, antichrist. ap-, i. Gr. fX7ro f from apanthropy, aphelion. 2. Lat. ad appeal, append. apo-, Gr. ano, apogee, apostate. Word Making. 61 ar-, as-, at-, aut-, auto-, be-, bene-, bi-, bin-, bis-, by-, cat-, cata-, circu-, circum-, cis-, CO-, col-, com-, comb-, con-, contra-, contro-, coun-, counter-, cu-, cur-, Lat. ad arrive, arrogant. Lat. ad ascend, assist. Arab, al assegay. Lat. ad attend, attest. A.-S. at, Eng. at atone, i. e., at one. Gr. avro?) self authentic. Gr. fxvro? autocrat, autograph. A.-S. be, a shortened form of bi, a prefix of very wide application. With verbs it intensifies or applies the action to some object, as in bedew, be- moan, benumb. With prepositions it has little force ; perhaps defines location more exactly, as in before, behind, beneath. Lat. bene, well benefit, benevolent. Gr. inij upon, over bishop from sniffKOTto'S. Lat. bi = dui, from duo, two biennium, bifurcated. Lat. binus, double binocular, binoxyde. Lat. bis, twice bissextile, bistort. Eng. by by-path, by-stander. Dan. by, a town by-law ; Dan. bylov, Icel. boejar log, local or municipal regulations. Gr. Kara, down, by, confronting catacoustics, cate- chise, catholic. Gr. Hard catalepsy, catastrophe. Lat. circum, around, about circuit, circulate. Lat. circum circumference, circumnavigate. Lat. cis, on this side cisalpine, cisatlantic. Lat. co, for con, a form of cum, with, together co- agulate. Lat. con = cum collateral, collocate. Lat. com = cum commingle, commotion. Lat. com combustion. Lat. con concatenation, concur. Lat. contra, against contradict, contravene. Lat. contro, against controvert. Lat. con council, counsellor. Fr. contre from Lat. contra countermand. Lat. con custom, from consuetudo. Lat. con curry, to work, or dress, i. e., hides, a horse, etc. NOTE. To curry favor is a corrup- 62 The English Language. tion of curry Fa-veil, that being an old English proper name for a horse. d-, Fr. de, of daffodil fieur d'asphodele. de-, i. Lat. de, down, from decapitate, degrade. 2. Fr. de, Old Fr. des, Lat. dis, asunder ; sometimes negative and oppositive, at other times intensive, or with a variety of meanings scarcely perceptible deform, defraud, desiccate, desolate, destroy. 3. Lat. dis defer, delay, deluge. demi-, Lat. dimidius, half demigod, demilune. des-, Fr. des, Lat. dis despatch, dessert. di-, i. Gr. dia, through, apart diaeresis, dioptric. 2. Gr. 6fe, twice diphthong, diptych, distich. 3. Lat. dis digress, dijudicate. 4. Lat. de, down distil. dia-, Gr. dia diameter, diaphanous. dif-, Lat. dis differ, diffuse. dis-, Lat. dis, often with an adversative signification dis- honor. dys-, Gr. 6v?j painful, difficult dyspepsia, dyspnoea. C-, i. Lat. e, out of evade, evolve, edict. 2. Fr. prosthetic, without meaning esquire, from Lat. scutum. 3. Du. ont, away elopement. CC-, Gr. KJ out of eccentric, ecstasy. ef-, Lat. ex, from, out of efflorescence, effrontery. el-, i. Gr. fry in ellipse. 2. Arab, al or el, the elixir, Arab, el iksir, the philoso- pher's stone. em-, i. Gr. v emphasis, empiric. 2. Fr. em, from Lat. in embroider, emboss. en-, i. Gr. ev encyclical, encyclopaedia, energy. 2. Fr. en, from Lat. in, negative = Eng. un enmity, endo-, Gr. svdov, within endogenous. enter-, Fr. entre, Lat. inter, among entertain. ento-, Gr. SVTOS, within entoblast, entozoon. ep-, Gr. snij upon ephemeral. epi-, Gr. en i epigram, epitaph. equ-, Lat. cequus equal, equanimity. CSO-, Gr. effGo, within esoteric. Word Making". i. eu-, ev-, ex-, exo-, extra-, for-, fore-, gain-, hemi-, hetero-, holo-, homo-, hyp-, hyper-, hypo-, i-, U-, im-, in-, inter-, intra-, intro-, iso-, juxta-, 1-, i. 2, mal-, i. Gr. ev, well, pleasantly eulogy, euphony. Gr. sv evangelist. Gr. e, from exodus, exotic. Lat. ^ H 3 Z T , t *V t* L c. + L Caph <^ 1 1 r J/ ^1 K K ^ Lamed. . . L t u ti z, ^ / L h Mem .... l *J 9 \*\J V } ^I/A M M to Nun 1 V L 5 L VI A/ * ^7 ; 7 / 7 / 1 A N j Samech . . ^ f ^ \ ty f H D Ayin , o o Ou) o O O y Pe -M ^ ^ L j n p n p j. " \J ' ' / 1 1 1 K Tsade s* r A A \ ^ 5? Koph.... &,

and J? ; that of th in thine by a d crossed, $. After the Norman Conquest the Saxon peculiarities gradually disappeared. Wen was replaced by two z/s, whence its modern name of double-u. The stricken d gave place either to a plain d, or to the runic thorn, which continued to hold its place until the middle of the fifteenth century. Its loss a really serious one was probably due to the early printers, whose types made for Latin and continental languages had no representative for this peculiar English sound. The art of making books by machinery introduced disorder in two other instances. The old Saxon g, besides the sound in go had, in middle English, two others, one of which is now entirely lost to the language, leading to further confusion. Sometimes alone, and always when combined with h, it had the value of the German g in Burg ; at other times it had the sound nearly of our initial y. For these two sounds the Saxon form had been preserved. The printers having no types for this character took others that seemed nearest in ap- pearance, not in value. They selected y to represent th, and z for either g or th. Thus it comes about that the people of the fifteenth and sixteenth centuries are represented in old books as if they said zour for your, zou for you, and ye, The Alphabet. 123 yat, yem instead of the, that, them, whereas they really spoke much as we do. 1 Every language requires an alphabet adapted to its own special system of sounds. Such alphabet, to be perfect, should have a separate character for every single sound, or shade of sound worth distinguishing, while no character should represent more than one. The letters should be easy to distinguish, and in writing should be easy to make ; and, combined in words, they should present a neat and ele- gant appearance. It may be safely affirmed that all these requirements have never yet been met. Our own alphabet offers little ground for complaint in regard to the last three, but in respect of the first two is sadly at fault. It is both defective and redundant. Each of the vowels represents several sounds. Ears and speakers differ, but a considerable number are easily distinguished. A has distinct sounds in amaranth, far, fall, wander, fare, and fame. E is variously pronounced in met, meet, there, and perfect, while in such words as permit and suffer it has an obscure sound scarcely distinguishable from a very short u. I has three well defined sounds, in pin, pine, and pique, to which some add pirn, girl, and alienate. O has four distinct sounds, in dot, dome, done, do, and a fifth tolerably distinct in wolf, to which some add a sixth in form. U is variously heard in rust, rule, full, mule, busy, and burial, to which some add turn. Y differs in pyx, pyet, myrtle, plenty, and yet, and some say in hyrse. W is treated as interchangeably vowel or consonant. I am unable to perceive in it anything but the sound of u in rule pronounced with varying degrees of quickness and force. To my ear the difference between the short sound of a vowel 1 This curious spelling was retained in a few words in Scotland as long as the Scottish dialect was spoken. As an example, spulzie or spuilzie (Scott's " Wa- verley," chap. 48.) was pronounced spool-ye. The late Saxon character for g had, however, begun to be mistaken for z by the middle of the fourteenth century. 124 The English Language. and the prolongation induced by a succeeding r between then and there, or but and burn is only one of duration. I see no greater difference than between the s in sun and that in hiss, or the / in pale and in pull. At the same time I dis- tinguish two shades of the long i or y in fine and in fire in fiy and in try. There is a kind of personal equation in this matter, as I sometimes detect differences unperceived by many others, and still oftener fail to perceive distinctions generally recog- nized. C performs triple duty in case, cease, and chess, to which we may add its use in spacious and machine. F is pronounced differently in ^and of. G is said to be hard in get, and soft in gem. H is employed in producing six sounds or combinations in hair, share, chair, sphere, there, and thorn. L and N, when combined with i, as in salient and lenient, yield peculiar sounds that may be attributed either to the consonant or the vowel. Pis employed to produce two different sounds in periphery. S stands for two sounds in dose and rose, and combined with i it yields two others really simple sounds in mission and fusion, the former of which is also produced by sh. T performs fourfold duties in time, thin, thine, and nation. Z in azimuth differs from the same letter in azure. Our alphabet is further defective in having no single char- acters for the sounds which we represent by sh, by th, and by the z in azure. It is redundant in that c, q, and x are superfluous. In call and cell, c could be replaced by k and s. In chair the ch is indeed a compound, but it is wrongly compounded, the real elements being / and the sound which we usually express by sh. It is obvious that q might always be replaced by either k or kw, and x by ks or gs. J is not redundant, since we have nothing to take its place, but it is a malformation. It should be d, followed by some character having the value of the French /, or the zhivete of the Russians. The Alphabet. 125 A great number of anomalies, as ph, ew in sew, eau in beauty, are not defects of the alphabet, but irrational spelling. Speech is produced by a mechanism combining the lead- ing features of the reed organ and the bagpipe. The lungs are the reservoir from which the air is urged through the flexible trachea, or windpipe, by the muscles of the chest and abdomen. The larynx and mouth, with their great powers of modulation, roughly correspond to the chanter of the Highland pipes, and the nasal passages may represent the accompanying drones. On the upper extremity of the windpipe is placed the valved box called the larynx the special organ of voice. It is composed essentially of four cartilages, four ligamentous bands, the lid called the epiglottis, and an exceedingly delicate arrangement of mus- cular and other tissues, too intricate to be described here. The lowest cartilage, forming the base of the larynx, is called the cricoid, meaning ring-shaped. It is but little modi- fied from the rings that compose the trachea. It is con- siderably higher behind than in front. Upon this rests the thyroid, or shield-shaped cartilage, composed of two plates, united at an acute angle in front, so that a horizontal section would resemble the letter v. The angle makes a carinate projection in front, easily felt by the hand in the upper part of the throat, and is popularly called Adam's apple, from a conceit that the forbidden fruit not only stuck in the throat of our first parent, but still inheres in all his descendants. The thyroid forms the greater part of the front and lateral walls of the larynx, but its ends do not meet posteriorly. Each extremity has two projections, termed horns, the one extending upward, the other downward. The lower ones articulate with the cricoid, leaving between the two carti- lages an open space in front and considerable freedom of motion. Upon the posterior part of the upper edge of the cricoid, articulate two small bodies called the arytcnoid, or ladle-shaped cartilages. By means of their controlling nerves and muscles, they admit of great celerity and delicacy of movement. From the last-named cartilages to the angle of 126 The English Language. the thyroid extend two pairs of ligamentous bands, one pair at a little distance above the other. The lower are known as the inferior or true vocal cords. They serve the same purpose as the reed or tongue in an organ pipe, and are the immediate determinants of the tone or pitch of the voice. When they are quiescent, the space between them, called the glottis, rima glottidis, or chink of the glottis, forms a slender triangle, from the angle of the thyroid to the two arytenoid cartilages ; but when they are tightened in pro- ducing the higher notes, there remains but a mere seam, not wider than the thickness of writing paper. In elocutionists, and still more in accomplished singers, these cords are ad- justed with marvellous quickness and delicacy to produce the various tones. Without them there is no voice. In whispering they do not vibrate. That is breath made articu- late, but not vocal, and may be imitated without allowing any air to pass through the glottis. M. Deleau illustrated this by passing a current of air through the nose into the pharynx by means of a rubber tube. If then the mouth assumed the successive positions necessary for articulation, whispering was heard without any action of the lungs or larynx. If at the same time vocal sound was uttered, speech and whispering were heard simultaneously from the same mouth. A little above the true are the false vocal cords, which do no not approach closely, or in themselves produce sound. Between them and the true there is on each side a concavity known as a pocket or ventricle of the larynx Morgagms ventricles, which seem to augment the voice by reverberation, as sound is intensified by partial inclosure in the fiddle or the drum. The piercing cries of the howling monkeys are due to extensions of these ven- tricles. At the superior margin of the larynx, held erect as a sen- tinel by elastic ligaments, stands the cartilaginous valve or cover called the epiglottis, which falls like an automatic draw- bridge when anything is to be swallowed. Above the larynx, the pharynx extends about four inches toward the base of the skull. It is a muscular sac, serving The Alphabet. 127 somewhat the same purpose as the air reservoir in a forcing pump, as well as that of the reverberating pipe of a wind in- strument. It is the common meeting-place of seven passages the aesophagus and windpipe below, the mouth anteriorly, and at its upper part the nostrils and Eustachian tubes that lead to the ears. From its extent and situation, its muscular structure and power of distention and contraction, the phar- ynx is of prime importance in giving character and volume to the voice. Separating the cavity of the pharynx from the mouth is the pendent curtain known as the soft palate, or velum palati. It is terminated below by the heart-shaped point called the uvula (little grape), is movable in speaking and swallowing, and does not at any time form a perfect closure. It rises and sinks in passing from one vocal sound to another, and aids in closing the passage to the nostrils. The more accessible parts that contribute to articulate speech the lips, teeth, tongue, roof of the mouth, and nose need not be described. Some writers discuss with great particularity positions assumed by the several organs in speaking, and even en- deavor to teach in that way the pronunciation of remote and unknown languages. But such directions, although sound in principle, are apt, in practice, to become unintelligible as soon as they become necessary, for the reason that a great part of these positions are out of sight and unknown. The reader will readily admit that the different articulations do depend upon the movements and positions of the organs above enumerated ; and can illustrate this to himself by pronouncing leisurely and carefully syllables beginning or ending with 0, oo, p, b, f, v, m, t, d, n, k, I, r, s, and th, and watching the positions assumed, so far as they can be observed. If now the mouth be opened moderately wide, and the tongue allowed to lie flat, so that the passage for the voice shall be unobstructed, and a vocal sound be uttered, it will be that of the a in far, or as I should prefer to say in alarm or Alabama. This may be regarded as the fundamental sound 128 The English Language. of human speech. It is the first utterance of infants, and abounds in the most primitive languages. It occurs twice in the simplest or ground form of most Hebrew verbs, and in Sanskrit is employed about half as often as all other sounds together. Next pronounce the word stop, allowing the pres- sure of the breath to cease before parting the lips ; otherwise it will be pronounced, stop-ih. The lips will be firmly closed and all utterance cut off. A opens the mouth wide ; / closes it completely. It might seem at first sight that all possible articulate sounds must lie between these extremes, and might be arranged in a series. And as every space is infi- nitely divisible, if only the divisions be infinitesimal, there is theoretically no limit to the number of intermediate sounds. The greater the number, however, the less dis- tinguishable. The number really found in different lan- guages and dialects, and in local and personal peculiarities of utterance, is very great. Mr. A. J. Ellis has devised an alphabet which he thinks capable of representing the sounds of all known languages. It consists of 270 letters ; and the Standard Alphabet of Lepsius contains 172 besides the tones of the Chinese and the clicks of the Hottentots. But may there rrot be other closures of the outlet for the voice besides the one above described ? In point of fact there are in English two others, observable in pronouncing pit and pick. One free passage, therefore, is contrasted with three complete closures, and articulate sounds might be exhibited in three lines diverging from a point in common. Such an arrangement is by no means new, and has been veiy fully presented by Professor Whitney in his Sanskrit Grammar, and other publications, and in a manner not widely different from the following. The letters within brackets represent sounds not now in the English language. The diacritic marks attached to the vowels are those used in Webster's Dictionary. The prolongation of a vowel is not regarded as a difference of sound. The Alphabet. X 2 3 4 vocal * 5 e 6 y 7 . ni >g 8 voiceless h 9 vocal zh 10 voiceless sh ii vocal [gh] 12 voiceless [ch] 13 vocal g 14 voiceless k palatals a a e a 6 a u e W rl n z s dh th d t linguals vowels w semi- vowels m nasals breathing ) sibilants v ) > spirants mutes conso- > nants P ) labials Words that will illustrate the sounds represented in the several lines are : 1 alarm 2 let, care, not, awe 3 bane, bun, bone 4 bit, bird, bull 5 be, German iiber, rule 6 yet, ray, lay, way 7 onion, singing, nun, mum 8 heigh-ho 9 azure, ooze 10 hashish, sister 11 Arabic Ghizeh, thy, vow 12 German noch, thin, fife 13 grog, deed, babe 14 kick, tight, peep The u in mule and the i in fire have been omitted because both are composite sounds. The former = yu, and the latter is composed of the first vowels on the first and fifth lines. It will be seen that the sounds of u in bun and i in bird occupy the centre of the triangle of vowels. They are indis- tinct sounds that have been compared to the gray among colors. All short, unaccented vowels tend to degenerate into these obscure sounds. Observe what there is of vowel sound 9 130 The English Language. in the last syllables of circular, paper, pallor, pillar, or the second of atrophy, harmony. As this is the first instance where a classification has been necessary, it may be as well to remark once for all that in nearly every attempt to classify a number of things some will be found to have claims on more than one class. The claims may be so nearly equal that to locate them anywhere will be an arbitrary sacrifice of principle to convenience. The oldest and most familiar division of letters is into vowels and consonants, but then come the semi-vowels between. F is properly enough classed with the labials. It is produced by emitting the breath through the slight chink left in bringing together the lower lip and the upper teeth. It is therefore nearly as much due to the teeth as to the lips. So the nasals are not produced by the nose, but only with its assistance. Indeed, very few are formed without the combined action of two or more organs. R may be pronounced as a pure lingual when the tongue is raised and made to vibrate with the passing vocal breath but does not touch the teeth or palate. The vowels are uninterrupted emissions of voice. The passage varies in form with each, but is unobstructed. The experiments made by Helmholtz and Koenig with graduated tuning-forks show that the vowels, as uttered by the same voice, are separated by regular musical intervals. As pronounced in North Germany, Koenig found the num- ber of vibrations to be approximately : u o a e i 450 900 i, 800 3,600 7,200. It is interesting to note that the old grammarians of India regarded o as a union of a and u, and e as a combination of a and i. (See Comptes Rendus, April 25, 1870.) The term consonant signifies sounding along with, as if in- capable of utterance without vowels ; but the sibilants cer- tainly need none, and scarcely do the spirants, and / and r are in some languages treated as vowels. We might say, then, generally that consonants are the result of arresting or The Alphabet. 131 obstructing the voice or breath ; but in that case what should we say of h, which is voiceless, and in the formation of which the breath is neither stopped nor impeded ? Still, it is true of all other consonants, and they naturally divide themselves into those in which there is complete closure and those that require only various degrees of obstruction. This is there- fore a division into momentary and continuous sounds, the former of which are sometimes called explosives, and some- times mutes. They are further distinguished into those requiring the exercise of the voice and those produced by mere voiceless breath. The former are often termed sonants and the latter surds. Strong and weak would seem to be more expressive. This distinction, in the case of the mutes, is that with the weak the closure is made or broken instan- taneously without any accompanying vocal murmur, while an initial strong mute is preceded, and a final one followed, by a brief resonance of the voice in the closed cavity of the mouth and pharynx. The want of a uniform alphabet has long been felt by mis- sionaries, travellers, and all who have to deal with languages that differ widely from common European standards. Among many attempts at a uniform system of writing second in importance only to a universal language that of A. J. Ellis, called paleotype, and the Standard Alphabet of Professor Lepsius of Berlin, seem at present to be re- garded with most favor. The former is used to exhibit the pronunciation in the great dictionary now in progress under the auspices of the English Philological Society, and the latter is employed by the latest edition of the " Encyclo- paedia Britannica" in transliterating foreign names. Both are chargeable with excessive refinement and hair-splitting, and neither of them has been made at all intelligible to the general reader. For these reasons, and because neither of them has yet obtained, or is at all certain to obtain, general acceptance, I shall not try to introduce either in this place, although aware that it is very annoying for a reader to find his page filled with marks to which he has no key. CHAPTER VI. GRIMM'S LAW. WHEN languages closely related are compared, many words are found in all that are much alike, both in form and meaning, yet seldom quite the same. The following is a very simple example : ITALIAN SPANISH. PORTUGUESE. FRENCH. uomo hombre homem homme cavallo caballo cavallo cheval terra tierra terra terre mano mano mao main. No one of these is an imitation of another. They are the common offspring of the Latin homo, a man ; caballus, a pack-horse ; terra, the earth ; and manus, a hand. There is sometimes a kind of method observable in this diversity, Thus we have GERMAN. ENGLISH. GERMAN. ENGLISH. Dorn thorn Thier deer Ding dick thing thick theuer Thiir dear door dun thin Thai dale durch through Theil deal Daum thumb Thau dew. In these and a multitude of similar examples, d in either language corresponds to th in the other. A somewhat similar mode of comparison has been applied to the whole Aryan family of languages, and especially in 132 Grimms Law. 133 their treatment of the instantaneous, or mute, consonants of the three series ending in k, t, and /. (See page 128), To these correspond the sonants g, d, b. Each of these six may be aspirated ; but all the aspirates are seldom found in any one language. The complete series would stand thus : k kh g gh t th d dh p ph b bh in which kh would be pronounced somewhat as in bulkhead, gh as in big-horn, etc. Of the members now known, Sanskrit alone has all the twelve sounds, but makes comparatively little use of the lighter aspirates kh, th, ph, while the graver gh, dh, and bh occur very often. Ancient Greek had the lighter set, x> ^ Pi Latin and the other members generally were without any. As Sanskrit is the best preserved, we may suppose that the common mother tongue of all had these twelve consonants. If now words containing all of these were inherited by each of the descendants, what were they to do with them, when they had dropped or forgotten part of the constituent sounds. Their case would not be very unlike the problem of placing twelve guests in eight or nine single beds. In fact, they would often put two in a bed, and perhaps sometimes on the principle of first come first served. The Greek, the second best appointed, would use x f r ^ an d g^, S f r th and dh, and cp tor ph and bh. Latin, having no aspirates, replaced them imperfectly by h and f, and occasionally by d and b. As an illustration of the effect we may take a word, or rather words, very familiar to the Latin scholar do, dare, dedi, datum, and its real or apparent compounds, ab-do, ad-do, circum-do, con-do, sub-do, etc. In the simple word do signifies give ; in the compounds cited, put or place. To all appearance they are the same word, and have been generally so regarded ; but they are really different, and correspond to the Sanskrit da, give, and dha, place. The Greek, though unable to preserve them perfectly, could still keep them distinct as didomi and tithemi. 134 The English Language. We may now do as the Greeks did, reduce the twelve consonants to nine, bearing in mind that the aspirates, kk, th, ph, were comparatively little used in Sanskrit, while gh, dh, and bh were very common. We will also premise that Gothic will be cited as the oldest representative of a numerous secondary family, embracing Frisian, Norse, Ice- landic, Danish, Swedish, Dutch, Saxon, English. The gen- eral system of corresponding sounds may then be represented in tabular form. The first line gives the sounds as spoken by the ancient Hindoos, the second those uttered by the Greeks in derivatives from the same words, the third those of the Latin tongue, etc. This table exhibits in condensed form what is known as Grimm's Law, which may be further illustrated by tracing a few words through their principal transformations. Sanskrit kal, to cover or hide ; Greek kalia, a shelter ; Latin celare, to conceal ; Irish calla, a hood ; Anglo-Saxon helan, to hide ; English hell, hole, heal, hull. Sansk. tan, to stretch ; Gr. tein-ein, to stretch ; Lat. ten-uis, stretched thin; Ir. tan-aigh ; Goth, than-jan ; Eng. thin ; Old High German diinni ; Modern H. G. diinn. Sansk. pad, go ; Gr. pod-, Lat. ped, a foot ; Goth. fot-u ; A.-S. fdt ; Eng. foot ; O. H. G. Vuoss ; Mod. Ger. fuss. Sansk. gan, to generate ; Gr. gen-os, kind ; gen-esis, origin ; gyn-e, a woman; Lat. gen-itor ; Irish gean ; Welsh gen-i ; Goth, kwens, kwein-s, kwin-o, a woman ; Icel. kon-a ; A.-S. cwe"n ; Eng. queen and quean ; also Eng. kin, kin-dred, kin-d ; O. H. G. khind, a child ; Mod. H. G. Kin-d. Sansk. dant, a tooth ; Gr. o-dont ; Lat. dent ; Welsh dant ; Goth, tunth ; Lithuanian dantis ; Old Saxon, Dutch, Dan., Swed., tand ; Icel. tonn, for tannr = tand-r ; A.-S. tdth ; Eng. tooth ; H. G. zahn. Few words in Sanskrit begin with b ; few in Saxon and none in Gothic, with the corresponding/. Sansk. ghama, the earth ; Gr. cham-ai, Lat. humi, on the ground ; Rus. zemlia, land ; Lat. homo, Goth, guma, a man, a son of earth ; A.-S. bryd-guma, a bridegroom. Grimms Law. 135 T3 bfl bo bjo bJD o k (X -^ a 43 43 4-1 V> 45 'rt i i *~ 4- >^> 4-> -4-> 4" ^ 4-" 4* X CT "J _V _V - v< n '^-^ *~^ *"^ h/} DJD ^S " " " ^ *" e bX N N "S * bXI . . bJJ 44 v^- ao b/3 45 45 rt "1 .a 'S : e c 5 o w 45 g J * tyl t^l in i) S 5 * M .S ^ ^ 45 * C 4J rt - Jo O JJ *O T3 T3 -g T3 j O O O 136 The English Language. Sansk. dkran, to sound ; Gr. thren-os ; Goth, drun-yas ; Icel. dryn-ja ; A.-S. drdn ; Eng. drone. Sansk. bku, to exist ; Gr. e-phu, he was ; Lat. fu-i, I was ; Welsh bu ; Irish bi ; Lith. bu-ti, to be; Goth, bau-an, to dwell ; A.-S. bed-n ; Eng. be ; Germ. bi-n. If we now take a somewhat narrower view of the subject, and confine ourselves a moment to the Teutonic sub-family, we shall find that it naturally falls into either two or three divisions, as we may regard them. High German ; Low German, Low Dutch or Platt Deutsch ; and Scandinavian. The Scandinavian branch comprises the Old Norse, as the pa-rent of the others, and the modern Icelandic, Danish, and Swedish. Their most marked peculiarities are a suffixed article and a reflexive form of the verb. The definite arti- cle is attached to the noun. Thus in Icelandic : madhr a man madhr-inn the man sonr a son sonr-inn the son vetr winter vetr-inn the winter hridh a storm hridh-in the storm holt a copsewood holt-itt the coppice The reflexive, or passive, form of the verb is made by appending a fragment of a pronoun signifying self Icel. gremja, to vex ; gremja-sk, to vex one's self, to be angry. In Danish nothing remains of the pronoun except the letter s. at give to give at give-s to be given at elske to love at elske-s to be loved at finde to find at finde-s to be found at faae to get at faae-s to be gotten at drive to drive at drive- s to be driven In other respects these languages belong to the Low Ger- man branch, and High German that is, the language of the interior, remote from the sea-coast, represented by the mod- ern literary German stands alone in the transmutation of sounds. This will be shown by exhibiting a few words in ist, Gothic; 2d, Danish; 3d, Swedish; 4th, Dutch; 5th, English ; 6th, German. Grimms Law. 137 I. 2. 3- 4- 5- 6. taihun ti tio tien ten zehn timr timmer timmer timmer timber Zimmer tindan tender tindra tender tinder ziinden tungo tunthus tunge tand tunga tand tong tand tongue tooth Zunge Zahn tvai to twa twee two zwei taikns tairan tegn taere tecken tara teeken tornen token tear Zeichen zerren threis thata thu thaursti than (Icel.) thrir " that " thu " thyrstr " thann (The sounds of th are wanting in all but Eng- lish and Icelandic. three that thou thirst than, then drei das Du Durst dann, denn thagks dags dails " thakkir dag (Dan.) deel dag del day deel thanks day deal Dank Tag Teil dal dal dal dal dale Thai dauhtar daur datter dor dotter d6rr dochter deur daughter door Tochter Tiire dragen dreiban drage drive draga drefva dreggen drijven drag drive tragen treiben It follows that of all the languages of this sub-family the literary German of to-day is the most remote from our own. The shifting of consonants is not confined to the initial sounds, as will be seen by the following additional exam- ples confined to the English and German. crib double stubble lead Krippe doppel Stoppel leiten ladder Leiter leaf Laub life Leben calf Kalb bridge ridge Briicke Rticken ship sweep water Schiff schweifen Wasser cat Katze malt Maltz salt Saltz earth Erde wether Widder give love geben Liebe CHAPTER VII. PRONUNCIATION AND SPELLING. IT is not the intention here to show how all words should be written and spoken. For information on these points, as on many others, the reader is referred to the dictionaries ; but as our spelling is admitted to abound in anomalies it is my purpose to make these a little more intelligible, by show- ing how some of them arose. This chapter has been headed advisedly, Pronunciation and Spelling, thus giving the spoken word precedence over the written, contrary to what I suppose to be the popular judgment. Languages are spoken long before they are writ- ten. Very few are written yet to any considerable extent. Were it possible to take an account of the words spoken and written on any one day, I doubt not the former would outnumber the latter a hundred to one. Writing is to speech as a portrait to the living face an attempt to repre- sent and perpetuate a perishable original. A man may not look like his portrait, but in that case which is correct ? Our aim should be not to pronounce as we spell, but to spell as we pronounce. Pronunciation and spelling agree and are consistent when the same written signs no matter what they may be al- ways represent the same sounds. They disagree when the same characters are assigned to different sounds, or different characters by turns to the same sounds. The English c and g represent at least two unlike sounds each, while f, ff, gh t and ph are used for the same sound. Twelve different com- binations are put for the vowel sound heard in peel, and eleven 138 Pronunciation and Spelling. 1 39 for that in no. In though half the word is in the position of a representative without a constituency, and the g in gaol is a peculiar and solitary exception. There is not a single letter in our alphabet that always stands for the same sound. R comes the nearest to it ; but in regard to that letter there is considerable diversity both in theory and practice, and it seems in danger of being entirely lost. In looking for the cause of these divergences we observe that people do not all pronounce alike, even when brought up amid the same surroundings. Minute peculiarities of organization combine with diversities of tastes, associations, and pursuits to produce dialectic differences of localities, families, classes, and trades. Webster's Dictionary gives a list of 1,275 words in the pronunciation of which authorities are not agreed ; and the differences among these experts are sometimes quite considerable. Again, our ears agree about as little as our tongues. If a foreigner were to recite to a hundred persons a sentence of say twenty words in his native tongue and manner, to which all were strangers, and they were to take the words down from his dictation, they would undoubtedly make a hundred discordant reports. In the third place, with a system of writing like ours, persons left to their own unaided judgment will differ much in their application of letters to express sounds. In examining many letters from various parts of Europe and America some of them wonderfully spelled, I have had the curiosity to note in how many ways the same word would be written, and I have found 210 variant attempts to write the single word com- mutation, all in good faith and under circumstances to put the writers on their best behavior. Moreover, when words are once committed to writing they remain in that form to be read for centuries, while living speech moves away, leaving them behind like old water- marks, showing the former course of an ever-shifting stream. And this is the principal cause of the divergence between pronunciation and spelling. When Butler celebrates the linguistic acquirements of Hudibras he represents that his hero 140 The English Language. " made some think when he did gabble, Th' had heard three laborers of Babel, Or Cerberus himself pronounce A leash of languages at once." These now seem poor rhymes ; but when men said Bab-el and pronunce they were perfect. Finally, if a set of characters were invented expressly for one language of few and simple sounds, the adaptation might be perfect till the language changed. This temporary suc- cess was probably attained by the Devanagari and Arabic alphabets and the Japanese and Cherokee syllabaries. An Indian chief is said to have written a letter in the last-named characters the day he first saw them, so easy it is to learn a system at once simple and self-consistent. But a bor- rowed alphabet, like a borrowed coat, is very apt to be a misfit. Now to apply these general considerations to our mother tongue, we learn that in Saxon England considerable diver- sities of speech prevailed, whence probably originated the present rustic dialects. Among these, through the political ascendancy of Wessex and the learning and patriotic labors of King Alfred, West-Saxon attained a temporary suprem- acy and has been regarded as the typical Anglo-Saxon. The Angles and Saxons had come from the shores of the North Sea as unlettered pagans. The religion and civilization of the Mediterranean coasts were brought to them by Christian missionaries. Their runes and beechen tablets where ex- changed for Roman letters, parchment, pens, and ink. In adapting the new alphabet to their wants they rejected k, q, and z, and they did not distinguish j from i or v from u. They retained an old rune to represent the sound of th in thin, and crossed a d (i$) for the th in then, as the Romans had no way of expressing these sounds. They introduced w with the same power as at present. Setting aside minor distinctions of different writers, the Anglo-Saxon alphabet represented the following sounds : Pronunciation and Spelling. 141 A, as in hart ; never as in hate or hare ; d, as in hall, or as in far. B, as in babe. C, " " case ; never as in cease or cheese. D, " " deed. E, " " men ; /, as a in mane ; indistinct when final, but not silent. F, as nearly v as possible. G, as in go, not as in gin ; like y before e and /. H, " " home, when initial ; in the middle or end of a syllable, like the German or Scotch ch the lost sound of English. I, as in tin ; t, as in machine. L, M, N, as at present. O, d, as in Soho ' ! P, as in puppet, but rarely beginning native words. R, trilled, or fully sounded. S, as s, when double, or when preceded or followed by c, p, or t ; otherwise like z. T, as at present. U, like the vowels in cuckoo. W, as at present never silent. X, seldom used, and then as a monogram for hs, or an ana- gram for sc. Y, y, as the French u and the German ii. Th (a single character) acquired the two values of ih in thin and in thine. D = th in thy. The chief diphthongs were : ae = a in care ; a, the same prolonged. ail, aw, ow = ow in now. ie, the same as / followed by a faint e. ea, ea, eo, / as it still is in German. That century was filled with disorders and civil wars, political and religious. It saw the acces- sion and expulsion of the House of Stuart, of which one king perished on the scaffold and another was driven with his sons forever from the inheritance of his fathers. It saw London twice desolated by plague and once by fire, a parliament turned out-of-doors by armed force, a commonwealth established and overthrown. It witnessed the rebellions of Montrose and Monmouth, the landing of a second William the Conqueror, and the fruit- less attempt of James to array the Irish against England and Scotland. So many great events had never before been crowded into a single century ; and the constant popular ferment had its effect upon the language of the people. The changes were so great that in many instances usage has receded from the extreme water-marks of that period. Among these were the sound of (sh), once heard in consume, pursue, sew, suit, supreme, but now met with only occasionally in Scotland and Ireland. There is an illustrative anecdote of a Scotch shopkeeper, who, worn out of patience, replied to a higgling customer : " Weel, weel, just shoot [suit] yersel, an' Al shoot mysel." To tell concisely the sound that s represents in each instance is very difficult. We are reminded again that usage is unconscious of any rule or principle, and that the systematizer can at best give but a semblance of regularity to the chaos of flying atoms. However, s initial is always (s) ; also when immediately before or after a weak consonant f, h, k,p, t, th in thin whiffs, goshawk, skips, speaks, starts, faiths ; in the prefix mis and the terminations ss and us 172 The English Language. mislay, darkness, circus, genius, famous. It is generally so in the prefix dis ; but usage is not quite uniform. I have been accustomed to except only disaster, discern, disease, dismal, dissolve. Even in these few the dis is not always from the same original. To these may be added, 1st, final as, except as, whereas, has, was, and a few foreign plurals; 2d, yes; 3d, final is, except is and his. Final s in all other connec- tions is generally (z) ; in a few words from the French it is silent. A following e sometimes preserves the sound of (s), which would otherwise be (z), as in asperse, expense. The final e does not generally save the (s) after a vowel lose, dispose, please, fuse, amuse (all verbs), but has that effect in base, dose, geese, lease, loose. We have thus a distinction, often but not always observed, that nouns or adjectives present a weak consonant where corresponding verbs have the strong. WEAK STRONG advice advise bath bathe belief believe brass braze calf calve close close device devise grass graze WEAK STRONG house house life live loss lose mouse mouse price prize rise rise thief thieve use use S = (sh) when preceded by a consonant and followed by ion, the i palatalizing the combination, as in diversion, expul- sion, mission, passion ; also when thus followed by a palatal- ized u sugar, censure, tonsure, sensual, fissure, pressure, impressure, sure, sumach ; and in the words nauseate, nause- ous, osseous, where the e following s has the palatalizing effect. When the termination sion, sure, etc., is preceded by an accented vowel, the s is (zh) instead of (sh) cohesion, contusion, explosion, disclosure, exposure, measure, pleasure, treasure. It has also the same value in combinations that are similar though not identical as crosier, osier, hosier, brasier, ambrosia, elysian, scission, abscission. Pronunciation and Spelling. 1 73 T when not modified by the following letter must have been always (t). But when a palatal vowel (u) or, what is the same thing, the semivowel y or (yu) follows, there is a tendency to palatalize the /, like c and s in the same situa- tion, to (sh). It never becomes (zh). This palatalization must have taken place as early as the fourteenth century, since Chaucer wrote indifferently discrecioun, discressioun, and discretions.. In regard to the termination ture, usage and opinion both differ, varying all the way from (naichuur) to (naitr). The first extreme is heard only from pedants on exhibition, the second from rustics. The admissible means are (naichur) and (naityur) of which I should prefer the latter ; but would say (temperatyuur) and (sepultyuur). From a want of suitable single characters, th is now in use for (th) and (dh) in thin and thine. This spelling superseded the Saxon thorn and crossed d in the fourteenth century. U and V are originally the same letter. From the Greek transcription of Roman names, it seems to have had in Latin the three values (uu, v, w). The first of these seems to have been the uniform Saxon sound, varied only by its greater or less duration. The Conquest, and prolonged intercourse with France introduced a great number of French words in which u was (ue). Out of these grew our present (yu), the vowel heard in use, unity, mule, pure. There are only three words in the language having this value of u that do not come to us through the French or the Latin. Of these hue and puke are native English, and emu, of unknown origin, was introduced into Europe by the Portuguese. It was when the long u had come generally to signify (ue) that the digraph ou was introduced to distinguish and pre- serve (uu). The change from (ue) to (yu) may be referred to the seventeenth century. U has now the six values, (yu) in mule, (uu) in rule, (u) in put, (u) in but, (i) in busy, and (w) in anguish, (u) and (i) may be termed the vanishing points to which all indistinct vocal sounds tend, for which compare the last syllables of ocean, further, kingdom, simpleton, captious, and the first 1 74 The English Language. syllables of perform, hirsute, congratulation. Even when not recognized as a separate vowel sound, it probably exists as a kind of residual product in many languages, and in my opinion is more frequent in French than in English, (u) has probably existed time out of mind along side of (uu), they being respectively the short and the long of the vowel, (ue) has gone entirely out of use in English. Persons in this country who try to learn German are apt to mistake it for (u) or (i). A trace of short (ue) remains in bury and busy, written in A.-S. with y, in early, middle, and modern English with u, but probably always with nearly the same sound, (u) was not recognized by orthoepists as an English sound until the middle of the seventeenth century, when it extended to many places now occupied by (u). V, as a consonant, seems to have been from a very early period always the same. There was originally only one form for u and v ; and when two simultaneous forms came into use in the twelfth or thirteenth century, it was not to dis- tinguish the vowel from the consonant. With some irregu- larity v was written at the beginning of a word, and u everywhere else. This usage continued into the seventeenth century, and was followed in King James's Bible. Sam Weller is a type of a class who find it difficult to pronounce an initial v. When I was young, there was in the neighbor- hood a man of that class from the old country whom the boys used to banter to say rapidly, " Veal, wine, vinegar, very good victuals, I vow." When thoroughly roused, the old man would run off in a string : " Yes, I can say ' weal, wine, winegar, wery good wit ties, I wow,' as fas' as any of ye." It can be little better than a superstition that prevents any word from ending with the letter v. The avoidance of cer- tain endings is curious. No word in English ends with /, except the names of a few foreign animals ; none in j, q, or v. Few native words end in o, still fewer in u, and none in a. As u is the last of the independent vowels, w being never used alone, and y only as the equivalent of i, this may be a Pronunciation and Spelling. 175 proper place to remark on the distinction of long and short vowels. In languages having simple scales of sounds the distinction might be merely one of duration, the sounds remaining the same, like musical notes differing only in time ; but in English the long and the short of any vowel are entirely different sounds. This will be seen by placing them in pairs, the long below the short. hat met bit not us hate mete bite note use Of most of our vowel sounds exact pairs cannot be made, unless by taking accented and unaccented syllables, as in notorious, in which we may call the first o short and the second long. The following are nearer making pairs, although as generally heard they are only approximations : LONG SHORT far fat mare met pane pa(ternal) fall folly LONG SHORT feel fill file fi(duciary) foal fo(ment) rule full but W. If the lips be protruded and brought near together, the vocal sound that can be produced is (uu), usually repre- sented by oo. If the lips be brought a little closer still, and if, on the instant that the voice issues, the organs of speech move to another position, the initial and instantaneous sound heard is w as a consonant. It has no perceptible duration, but is a mere starting-point, like/ in part. In our language it must be followed by a vowel. It is certainly possible, but not easy, to pronounce r after it, on ac- count of the vocal character of that letter. L is still more difficult. Wh has been the subject of much dispute. In my judg- ment it is (hw). Bearing in mind that h is not a vocal sound but a mere breathing, try this experiment. Pronounce sep- arately, with intervals between, h oo en ; then diminish the 1 76 The English Language. intervals until they disappear, and you will have a fair approximation to when. Next try the order, oo h en. Thirdly, place the lips in the position of w, and say when slowly and carefully. You will find that there is first the voiceless breathing, which ceases the instant the vocal w, or (uu), is reached ; that passes instantly into e, followed by n. To speak of wh as a simple but peculiar consonant partaking of the properties of w and h seems to me little better than to call it a univocal compound of sound and silence. As a vowel w = (uu), and as such it occurs alone in Welsh, but never in English, where it is always preceded by a, e, or o. After a it merely indicates one sound of that vowel. After e, omitting the exceptions cited under that letter, it is (uu). I thus construe ewe. The last letter is final silent e ; the first gives the palatal sound expressed by y, leaving w = (uu). Latin and the languages derived from it are without the letter w, but have the sound, which most of them express, as in English, by u, as in the Latin anguis, Italian guado, Spanish igual, English languish, assuage. There is a pronunciation remarked on in England as pecul- iarly American, which culminates in, but is not confined to, the word whole. It consists in making the o " long " in sound but short in time, and dwelling upon the / as if it were doubled, thus differing from hole in making the o shorter and doubling the /. The word then insensibly becomes hull. It ought to differ from hole only in having the o longer. A still worse pronunciation is appearing among some of those who insist on speaking as they spell, in trying to make the w audible. This produces a most uncouth and barbarous vocable for sake of a written letter that is no part of the original word. In all the equivalents except the English the w is wanting. It appears to have crept in from some local dialect about the year 1500, and attached itself to a number of words in the manner of the cockney h. In some words, as whole, whore, and whoop, it has been retained in the writ- ten but not in the spoken language. One and once present the w to the ear but not to the eye ; and from hoot and hot, Pronunciation and Spelling. 177 hoary and holy, it has disappeared altogether. As to wholly, which Walker very justly thought should be written wholely to correspond with solely, I have always been accustomed so to pronounce it. X is a compound equivalent to ks or gs. It begins no English word, and is followed only by c, ch, h, /, q, s, t, or a vowel. It is (ks) when final, when accented, or when fol- lowed by any of the consonants named except h. As x = ks, the element s becomes (sh) in situations where a single s would undergo the same change anxious, fluxion. Y originated, as we have f seen, among the Greeks, with whom it probably had nearly the value of (ue), which is peculiarly liable to pass into (ii, i), When the Romans be- gan to study Greek as a polite language, they distinguished this letter from all their own vowels by introducing it as a new character ; and the French and Spaniards still call it the Greek i. As a vowel its various shades of sound have ranged between the extremes i and u. The sound (ue) is one that the English early sought to avoid. It seems to have disap- peared in the thirteenth century, to have returned again with the influx of French words in the fourteenth, written u or eu, only to disappear and become unpronounceable in the seventeenth. In Scotland, at least near the Border, floor may perhaps still be called (fluer), but farther north will be found various substitutes as (fluur, flyuur, fleer). But when the sound first died out it parted company forever with the letter y, which then became indistinguishable from i. " I schal ryse, and I schal go to my fadir, and I schal seie to him, Fadir, I have synned agens heuene, and bifore thee ; Now I am not worthi to be clepid thi sone, make me as oon of thi hyrid men." LUKE xv., 18, by WYCLIFFE. Gradually custom assigned y to certain places and i to others, so that they are never interchangeable. F" is a vowel equal to i except when initial, in which latter case it has always the force of a consonant, and is to (ee) as w to (uu). 1 78 The English Language. PRINCIPLES OF SPELLING. 1. Every word of considerable length can easily be divided into a number of parts, as cir-cum-loc-u-to-ry. Each of these parts is called a syllable, from a Greek original signifying taken together, and consists of a single vowel sound, together with such letters as are pronounced along with it, without the aid of any other vowel sound. It may be a single vowel alone, or a vowel sound represented by several written char- acters, preceded or followed by consonants, or even by vowel characters not sounded. A, an, no, now, stone, view, thieves, are examples. 2. An open syllable is one ending with a vowel, one that ends with a consonant is called a closed syllable. Under the head of word-making I have shown reasons for believing that all syllables were originally open, and that a closed syl- lable is the remnant of a dissyllable that has lost the last vowel. 3. Our language, like many others, has a decided prefer- ence still only a preference for open syllables. Any Eng- lish-speaking person seeing for the first time the words carat, pater, would naturally divide them ca-rat, pa-ter, and not as they should be, car-at, pat-er. 4. The tendency is to give a prolonged, slender sound to the vowel of an open syllable. To this there are many exceptions and limitations. A final a is always short, hence the article a is never what is called long. / and a end no native English word, unless they can be said to do so when standing alone. 5. The vowel of a closed syllable is preferably short. It is uniformly so only when a single vowel is followed by a single consonant. The vowels of closed syllables are pro- longed in a great variety of ways, many of which are mere contrivances for showing that the vowel sound is long. Of these the first in rank is the silent final e, which makes the difference of bat and bate, bit and bite. Then combinations of vowels, whether pronounced in whole or only in part, are generally, but not universally, long. There are also a few instances where i and o followed by two, or even three, con- sonants are long bind, blind, behind, find, grind, hind, kind, Pronunciation and Spelling. 1 79 mind, wind, pint, bolt, colt, dolt, jolt, molt, bold, cold, fold, gold, hold, mold, old, sold, told, wold, fort, port. Finally i followed by gk silent is long sigh, fight, lights. The lengthening of the i is a compensation for the loss of gh. 6. The termination ed of verbs is written in full, however pronounced, as in landed, stamped, hushed, snuffed, sighed. There are, as usual, a few exceptions. We write heard, not heared, and laid, paid, said, and sometimes staid instead of layed, payed, sayed, and stayed. Made, shortened from maked, is disguised almost beyond recognition. When a verb already ends with an e, one e is omitted live, lived, smile, smiled. One e is also omitted with the termination er, and generally before ing. It is retained in dyeing, singeing, swingeing, tingeing, to distinguish them from dying, singing, swinging, tinging, and in hoeing corn, shoeing horses, and toeing a mark. In general, a silent final e is dropped be- fore all suffixes beginning with a vowel : Fine, fined, finer, finery, finest, fining, finish, finable j bride, bridal; guide, guidance; plume, plumage; grieve, grievance; move, movable ; force, forcible ; ice, icy ; true, truism. 7. A silent final e is generally retained before a suffix beginning with a consonant, or when necessary to preserve the sound of the original word peaceful, peaceable, change- able, courageous, mortgageor. It is sometimes omitted when it immediately follows the main vowel due, duly, duty, and, contrary to the soundest analogy in such words as judgment and lodgment. Truly, like truth and trust, is consistent, being older than the present spelling of true. Wisdom is not from the modern word wise. It is Saxon unchanged. 8. Any monosyllable or accented final syllable, having a single vowel and closed by a single consonant, doubles that consonant before a suffix beginning with a vowel : Fit, fitted, fitter, fittest, fitting. Red, redden, redder, reddest, reddish. Rid, riddance, ridder. pi g, Pigged, pigging, piggery, piggish. Confer, conferred, conferree, conferrer, conf errant, conferring. i8o The English Language. If the consonant were not doubled, the preference for open syllables would cause the first line of this example to be read : fit, fi-ted, fi-ter, fi-test, fi-ting. 9. When the last syllable is not accented usage ceases to be uniform. a. The termination ly in civilly, morally, etc., is not a doubling of the last letter, but the same ly as in comely. b. Final c adds a k, and g a second g, to preserve the hard sound before suffixes beginning with e, i, or y. c. As an accent on the last syllable causes a doubling, so if a word consists of two nearly equal syllables, the final consonant is doubled hob-nobbing, kid-napping. This is the only reason that can be assigned for doubling the/ in wor- shipper, as is most frequently done. If such words as hard- ship, flag-ship, war-ship, could admit of these terminations, the last letter would probably be doubled. It remains single in filliped, galloped, and walloped. We can only sug- gest that lip and lop are elements of less volume than ship. d. The following list contains most of the words of two syllables that are written sometimes with and sometimes without doubling the last letter before the suffixes ed, er, est, ing, ish, ist : anvil cudgel hatchel pencil snivel apparel dial hovel peril stencil barrel dishevel jewel pistol swivel bevel dowel kennel pommel tassel bias drivel kernel postil teasel bowel duel label pupil tinsel brothel enamel laurel quarrel tonsil bushel equal level ravel towel cancel flannel libel revel trammel carol focus marshal rival travel cavil fuel marvel rowel trowel channel funnel medal sandal tunnel chapel gambol metal shovel vial chisel gravel model shrivel victual counsel grovel panel sibyl crenel- handsel parcel signal Pronunciation and Spelling. 1 8 1 Dishevel and enamel, although trisyllables, having the accent on the second syllable, follow the analogy of dissyl- lables. Of the whole number 76 end in / and 2 in s, letters very apt to be doubled at the end of words. More- over a single s is liable to be pronounced as z. Other suf- fixes, not originally English, often follow the analogy of the original words from which these are derived, but not with entire regularity. Perry in England and Webster in America are the principal advocates of the single letter. In this country probably the majority follow them in most instances, while the opposite practice prevails in England ; but no one seems to be entirely uniform and consistent. 10. Trisyllables of similar endings, accented on the first syllable, and having a consequent secondary accent on the last, should, according to the widest analogy, double the last letter, but only a few of them, as handicap, manumit, ricochet, having the second accent well marked, do so in- variably. Reason and analogy are allowed to count for little. A few are variously spelled by different writers, of which carburet, sulphuret, pedicel, sentinel, and hospital, need no special remark. Parallel scarcely ever doubles the final letter for no other visible reason than to avoid four z's. Bishop Hall has even unparallelable. Most authors write compromised and benefited. As mere English words there is no reason for the difference. One may possibly be found in their derivation from the Fr. compromettre and bienfait. 11. A word ending with a doubled consonant, other than /, retains both letters before suffixes, ebb-ing, add-ed, odd-ly, stiff-ness, embarrass-ment. Usage is not uniform in regard to a doubled / before a consonant : dull-ness dul-ness enthrall-ment enthral-ment enroll-ment enrol-ment thrall-dom thral-dom full-ness ful-ness skill-ful skil-ful fulfill-ment fulfil-ment will-ful wil-ful install-ment instal-ment The shorter form prevails in England ; the longer among those Americans who accept Webster as an authority, 1 82 The English Language. holding that two /'s are necessary to preserve the sound after a and o. Full and fill are especially liable to be cur- tailed. In general full retains both /'s in compound words as full-blown, full-sailed, but is liable to drop one before consonantal suffixes. Some write fulfill and others fulfil. but no one writes fullfill. 12. Words ending in a doubled consonant generally retain it when they are lengthened by prefixes. boss emboss sell undersell buff rebuff tell foretell fall befall thrall enthrall roll enroll staff tipstaff It is the practice in England, and to some extent in America, to omit the last / in such cases, especially in distil and instil. Many words compounded with all, well, and mass omit the last of the two consonants : almighty, almost, alone, already, also, although, altogether, always, withal, where- withal, welcome, welfare, Candlemas, Christmas, Lammas, Michaelmas. All compounds of which full is the last part drop the last /, artful, hopeful, etc. We also write until, not untill. Mas is really a shortening, but the reason of the others is that the old spellings were al, ful, til, and wel, and the compound words were formed before the adoption of the modern spellings of all, full, till, and well. 13. Some combinations of letters are avoided. A, i,j, k, u, v, x, and y are never doubled, and w and h only in glow- worm, slow-worm, with-hold, and a few imperfect combina- tions not yet fairly consolidated into words rough-hew, high-heeled, etc. Q never occurs without being followed by u. /never precedes the closely allied letters/ and y. Some combinations are preserved that cannot be pronounced in full as phthisic, giaour, caoutchouc, but then we skip the hard parts. 14. Up to the fifteenth century i and y were used indis- criminately, and there are a few remains of their tenancy in common, in which we sometimes meet with the one and Pronunciation and Spelling. 183 sometimes the other hythe or hithe, tryst or trist, gyves or gives, mystery or mistery? syllabub or sillabub, cypher or cipher all old words except the last, and seldom met with. Y retains sole possession in scythe and byre. At present this letter has its position defined with more than usual regu- larity. a. As a consonant it is the first letter of a word, or of a part elsewhere independent be-yond, hal-yard, steel-yard. b. As a vowel it is final, with or without a preceding e, and takes the place of a number of terminations, Latin ius, ia, turn, as, as in amatory, controversy, estuary, civility ; Fr. and Eng. e, ee, i, ie, in ditty, dainty, majesty, cony ; A.-S. ic and ig in homely, happy. As a final too it is a diminutive express- ing affection, chicky, birdy, kitty, pussy, Bobby, returning to ie in Annie, Jennie, Nellie. c. Y forms a digraph or a diphthong after a, e, o, and rarely u, in any part of a word defraying, abeyance, employ, buying. d. It is vjery common as a vowel in words derived from the Greek symbol, amethyst, baryta. e. The following words, although not of Greek origin, are written with y from having passed through, or from being supposed to be in some way connected with, that language beryl, gypsey, gypsum, hyssop, lachrymal, lymph, papyrus, syca- mine, sylvan. f. The spelling of dye and lye serve to distinguish them from die and lie. The y in rye has remained since Saxon times. g. The y in nylgau and typhoon is due to the whim of those who transliterated these Eastern words into European characters. h. Words of more than one syllable, ending in y without another vowel before it, change it to i before suffixes begin- ning with a, e, or a consonant, but retain it before i to avoid the doubling of that letter holy, holier, holiest ; carry, car- 1 Meaning an art, trade, dramatic performance, from Lat. minisierium, and erroneously supposed to be the same as the Greek mystery, an esoteric or secret doctrine. 184 The English Language. riage, carries, carried, carrier, carrying ; bury, burial. In the few words of one syllable usage is much at sea. Verbs fol- low the analogy of dissyllables try, tries, tried, trial, trying. Nouns from these verbs, denoting the doer, admit of both forms crier and cryer, drier and dryer, flier and flyer, pliers and plyers with a preference in this country for i. Other nouns are regular. The adjectives shy, sly, spry, and wry retain the^ throughout. On the other hand, monosyllables in ie necessarily change the i into y before i, die, dies, died, dying. 15. There are now about 1,000 adjectives in the language ending in -able or -idle, and a very few in -eble, -oble and -uble. The primary meaning in this class is capability of undergo- ing some action. Movable, fordable, fusible signify suscepti- ble of being moved, forded, or fused. They are either from Latin verbal adjectives in -abilis, -ebilis, -ibilis, -obilis, -ubilis, or imitations of such. To the question which vowel to use the following is an answer in part. a. The termination when applied to words not derived directly or indirectly from Latin is -able answer-able, bear- able, drink-able, eat-able, fore know-able, unquench-able. As this implies a knowledge of Latin it is rather a reason than a rule of practice. b. Words, from whatever source, that have been long and commonly used in the language take -able agree-able, bail- able, comfort-able, service-able, unconquer-able. This leaves room for doubt and irregularity, as it is a question of degree how long and how much a word has been in use. Some are written in two ways, as add-able and add-ible, convers-able and convers-ible, refer-able and referr-ible, the first forms following a direct English, and the latter an indirect Latin, analogy. In a few instances there are two different words from the same source defend-able and defens-ible. c. When a Latin original is kept in view, the verbal ad- jectives in -abilis and -ibilis, if there be any, are followed. Latin verbs are divided into four classes according to the last letter of the essential part of the word. If that be a, the derived adjective ends in -abilis ; if anything else, in Pronunciation and Spelling. 185 -ibilis. If there be no such Latin adjective, then the Eng- lish one follows the characteristic of the verb. About two thirds of the entire class end in -able. 1 6. There are a number of words ending in -nee and -nt, preceded by a or e, and there is sometimes an uncertainty which of the two should be used. These terminations be- long properly to Latin participles and verbal nouns, and come to us either directly or through the French. If the characteristic of the Latin verb be a, the result is -ance and -ant, the first of which makes a noun, the second an adjective. Thus from the Latin abund-a-re, to abound, are formed : Latin, abundantia English, abundance abundant-s abundant So verbs whose characteristics are e or / yield forms that are regular : sil-e-re silence, silent sal-i-re salience, salient conven-i-re convenience, convenient The still greater number characterized by consonants ought regularly to furnish -ence and -ent : string-'-re stringence, stringent solv-'-re solvency, solvent resid-'-re residence, resident but usage, formerly very unsettled, is still by no means con- sistent. We have from consonant verbs affiance, affiant, ascendant, defendant, attendance, attendant, repentance, re- pentant, and many others. This irregularity is due to the passage of these words through the French language. A few words have both forms, as confidence, confident, and con- fidant ; dependence, dependent, and dependant. In such cases the form in -ant is treated as a noun, and that in -ent as an adjective. 17. A number of verbs ending in -ize are formed either from or in imitation of Greek verbs in -izo. Agonize, baptize, 1 86 The English Language. dogmatize, ostracize, syllogize, are examples of those from similar Greek words. Eulogize, analyze, paralyze, are from Greek originals not ending in -izo. Civilize, detonize,judaize, mesmerize, naturalize, realize, spiritualize, are various imita- tions. The manner of writing these words has never been uniform. We generally find catechise and exorcise, although from Greek originals in -izo. In this country the greater part are written with -ize ; the following are the principal exceptions, of which only two are from the Greek : advertise comprise divertise misprise advise criticise emprise premise affranchise compromise enfranchise reprise apprise demise enterprise revise catechise despise exercise supervise chastise devise exorcise surmise circumcise disguise improvise surprise In England the tendency is to write them all with -ise. 1 8. Words introduced into the language from any source are treated for a time as strangers, and allowed to retain something of their foreign look and sound. This is particu- larly true of the Italian terms relating to music and painting and the French phrases used by people of fashion. Although these foreign features are gradually effaced by long and common use, they are still one of the two principal causes of the discordance between pronunciation and spelling, of which asthma, bayou, Bootes, Canaanite, catsup, chapeau, cognac, corps, ret, schorl, are examples. When a word comes through one language, from another, it is sometimes a question which of the two should deter- mine its form. The largest element in our language is Latin, that has reached us through a transforming French medium. In general the French is followed, especially in England ; but in this country there is sometimes a disposi- tion to revert to the earlier type. One class o'f more than a hundred words begin with French en or Latin in, like enclose, or inclose, enquire or inquire. They seem to be the sport of chance, yet in most instances a preference is shown for one Pronunciation and Spelling. 187 form rather than the other, favoring e in England, and in this country inclining to i. There is another class with numerous imitations, the origi- nal members of which ended in Latin in -or, changed to -eur in French, adopted into English as -our, and now in America reverting to the orginal -or. FRENCH ENGLISH AMERICAN candeur candour candor coleur colour color doleur dolour dolor erreur errour error facteur factour factor faveur favour favor ferveur fervour fervor honneur honour honor odeur odour odor 19. In all cases of doubt consult the dictionaries. If they be hopelessly at variance, the only resource left to the stu- dent is to observe the spelling of the best recent writers, and exercise his own judgment. ACCENT. In a word of more than one syllable some one is uttered more forcibly than the others. That additional force or stress is what is here called accent. It is much more marked in some languages than in others. In those that are faintly accented the words and syllables ripple along in what seems to us a lazy droning fashion without spirit or force. It matters not how long the words may be, as the oral speech seems scarcely divisible into words, but to be drawn along before us like the successive links of a chain, or a string of uniform beads. On the other hand, our language is very strongly accented, and seems to foreigners irregular, jerky, sputtering, and quarrelsome. We think that the strength of our lan- guage, as the emphasis of earnest speech, is laid on the accented syllables. 1 88 The English Language. In many languages the accent is always near the end of the word, in others near the beginning, while a few, like ours, seem to distribute it with a show of impartiality. All attempts to assign rules for the place of the accent in Eng- lish only serve to render the subject hopelessly intricate and confused. Still some leading tendencies may be discovered, and where the confusion is inextricable the cause of that confusion may be found. The principal cause is the compo- site character of the language. There is first what may be called the native element either Anglo-Saxon in origin, or developed in English at an early period. About the pro- nunciation of this portion all are agreed. Then there are the multitude of words from Greek, Latin, and foreign languages, altogether strange at first to the body of the people. Let us then in the first place see if we can discover any principle regulating the accent in the native part of the lan- guage. I remark in passing that the place of the accent is usually denoted in English books by the mark (') thus ty'rant, embit'ter, require' ; that its place is reckoned from the last, or ultimate, syllable ; that the one next to the last is called the penultimate, or, for shortness, penult ; the sec- ond from the last, antepenult, and the one before that,/raz#- tepenultimate. Beginning then with native words of two syllables, as the simplest, we readily discover three tenden- cies, quite independent, and sometimes competing with each other. The first is to accent the principal syllable, where there is much disparity. It could scarcely be otherwise. Awake, at hirst, bestride must be accented on the second syllable, if at all, and stranger, proudly, on the first. The second tendency is to accent the first syllable, if there be no marked inequality. Thus pairs of syllables that are mere variants of each other, as hob-nob, criss-cross, hodge-podge, hum-drum, pic-nic, see-saw, are all accented on the first sylla- ble. So are chaffinch, chilblain, distaff, daylight, foresight, larboard, starboard, lukewarm, mermaid, nothing, onslaught, and many others, in which the second syllable by itself requires the most effort to articulate. I have before me a list of the principal independent native dissyllables in the Pronunciation and Spelling. 1 89 language, amounting to 955, of which 763 are accented on the first syllable. If the derivatives formed by the suffixes ed, er, es, esf, ish, ist, ly, etc., were counted, several thousands would be added to the list of penultimate accents, and not one to the ultimate. As it is, of the 192 words accented on the last syllable, 104 are made with the weak prefixes a and be abide, become, etc. The third tendency is to accent verbs on the last syllable. This will be best shown by contrasting nouns and verbs in which the first syllable is the same : fore'-foot fore-go' out'-break out-bid' fore'-ground fore-know' out'-come out-do' fore'-hand fore-arm' out'-let out-go' fore'-head fore-cast' out'-look out-grow' fore'-land fore-show' out'-house out-live' fore'man fore-stall' out'-law out-run' fore'-sight fore-tell' out'-works out-wit' This last tendency sufficiently accounts for our meeting with words of foreign origin, like contract, which when used as nouns are accented on the first syllable, and on the last when they are verbs. The final accents are nearly all due to weak prefixes or the use of the words as verbs. Passing next to native English trisyllables, I find in Skeat's Dictionary 125. The number in the largest American dic- tionaries is much greater by including those formed by adding suffixes, which have the effect, as before, of throw- ing the accent back towards the beginning. Of the 125, inasmuch, insomuch, and upsidedown may be left out, as being phrases rather than words. There would then remain 58 accented on the first syllable, 22 on the second, and 42 on the last. Of this last division, 36 are verbs ; overworn, overwrought, and unaneled are participles, with ultimate accent because of their verbal character. The remaining three words, overhead, overmuch, and overivise, have the character of composite phrases. There is not a noun in the list of 42. Of the 22 accented on the middle syllable, 10 are verbs, 7 have the weak first syllables a-, be-, to-. The other 5, like almighty, already, are as easily explained, with 190 The English Language* the exception, perhaps, of newfangled. There is but one noun in the class. Of the 58 accented on the first or ante- penultimate syllable, there is but one verb caterwaul, in which the balance is nearly even between the first and last. By general analogy, it should be ultimate. That an excess of volume in the first syllable is not essential, will be evident from such words as alderman, bedridden, didapper, emberdays, forefather, godmother, honeycomb, indwelling, offscouring. The tendencies of English pronunciation then are such that, of words of three nearly equal syllables, all verbs would be accented on the last syllable and all the others on the first. Of native words of more than three syllables, those accented on the first are about three times as numerous as those accented on the second, and these latter bear the same ratio to those having the accent on the third, while ultimate accent is limited to a very small number of verbs. This tendency to accent the first syllable produces some curious results, both in native and adopted words. The following are found in the dictionaries accented on the first syllable : almainrivets caterpillar-catcher ganglionary alveolary disciplinableness lachrymatory bluestockingness disputableness mainpernable calculatory excellency calipercompasses explicableness In very long words, or words having two syllables of con- siderable volume with a faint one between them, there is a secondary accent, or even two or more, besides the principal one hy'drocyan'ic, im' mate' rial' ity. Besides the three tendencies above described, there is a fourth, affecting words from learned and foreign languages. So long as they are used only by scholars and specialists, there is an effort to keep up their foreign pronunciation. But they gradually percolate through the masses, who naturally tend to assimilate them to their native speech. In this way there may be two pronunciations current at the same time. There is a drug much used in the South and Pronunciation and Spelling. 191 West that fashionable people call kin-een' , but which the dictionaries and common folks call qui'nine or quin'ine. So scholars say abdo'men and opponent, but the majority are strongly inclined to say ab'domen and opponent. The battle is fought over every dubious word separately, with victory in the long run generally with the majority. It is obvious that the result, even when reached, must be very irregular. Words that have been long in common use are accented like original English. Armor, equal, rapid, prophet are accented on the first syllable because that is the general habit ; attend, impend, combine are accented on the last, for the double rea- son that they are verbs and the last syllable is the heaviest. The English tendency to accent the first syllable has sub- stituted an antepenult accent for a penult in many words, and in others the change is in progress, there being one pronunciation by scholars and another by people in general. ora'tor or'ator lyce'um ly'ceum sena'tor sen'ator anemo'ne anem'one cica'trix cic'atrix elegi'ac ele'giac abdo'men ab'domen eurocly'don euroc'lydon minis'ter minister charac'ter char'acter pletho'ra pleth'ora dysen'tery dys'entery cogno'men cog'nomen panthe'on pan'theon muse'um mu'seum thea'tre the'atre umbili'cus umbil'icus An example of this gradual naturalization is furnished by a group of very recent botanical terms ending in phyllum or phyllous Greek and " Piers Plowman," 1362. After that time eves was Nouns. 269 mistaken for a plural without a singular. The original form is shown by the compounds, which are made with eaves and not eave eaves-board, eaves-dropper. Folk a common collective word for an indefinite number of persons, for the community in general. Used with a plural verb. In England it is more common to say folk are, and in America folks are ; but the usage is not uniform in either country. Here folk seems rather affected, and folks is no novelty in England. " Yet merry folks who want by chance A pair to make a country dance, Call the old house-keeper and get her To fill a place for want of better." SWIFT : " Stella's Birth Day." " Necessity and a little common sense produced all the com- mon arts, which the plain folks who practised them were not idle enough to record." WALPOLE'S " Anecdotes of Painting," chap. 5. Gallows strictly plural of a singular gallow, in use down to the middle of the sixteenth century. Kington Oliphant ("New English," chap. 2) quotes, "4 payre of galowys" from the early part of the fifteenth century. This instru- ment of execution was next called simply the gallows, and on the supposition that the name was singular a duplicate plural, gallowses, was formed. " The fear of gallowses and ropes Before their eyes might reconcile Their animosities a while." " Hudibras," part in., ii., 716. I have also met with gallowses in the Spectator, but have mislaid the reference. All the extant compounds are made with gallows none with gallow. Genius has a Latin pluralgenn for the creatures of Eastern fable and story, and English geniuses for persons of rare mental gifts. Horse. We use horse for one animal, horses for several, and again horse for a body of cavalry or troops on horseback. 270 English Grammar. Horse and foot, a phrase that came into use in the early part of the seventeenth century, may be an abbreviation for horsemen and footmen. The collective singular, horse, is not generally applied to animals without riders, yet Byron's " Mazeppa " was met by " A thousand horse, and none to ride." Index has a Latin plural, indices, for the characters that distinguish algebraic powers and roots, and the English indexes, when the word is employed otherwise. Madam, Fr. Madame, ma dame, Latin mea domina, Eng. my lady. At present no other plural is in use than Mesdames, which is sadly out of tune with our mother tongue. As Madam is no longer pure French, I think Madams not only permissible but preferable. Means Old French meien, Mod. Fr. moyen, Latin me- dium from medius, middle ; used as a noun or an adjective in reference to a point between extremes, as mean time, mean annual temperature, a safe mean between extremes. It next gets the sense of an intermediate agency by the aid of which anything is done. " The virtuous conversation of Christians was a mean to work the conversion of the heathen." HOOKER. But in this sense it generally takes the plural form. The word occurs thirty-two times in the Bible, but always as means. It takes singular or plural verbs and pronouns indifferently. "By this means thou shalt have no part on this side the river." EZRA iv., 16. "By these means, the queen had collected an army twenty thousand strong." HUME'S " Hist, of England," chap. xxi. Memorandum has two plurals, Lat. and Eng. memoranda and memorandums. A useful distinction is sometimes made by confining the former to a number of notes taken collec- tively, and the latter to notes that are separate and inde- pendent. For example : " He shewed me a paper containing Nouns. 271 memoranda relating to," etc. " He drew from his pocket a number of memorandums" Mister a thin and meagre degradation of Master, sup- posed to be formed in imitation of Mistress, which latter is from maister-ess. It is rarely, if at all, met with in the plural, for which the French Messieurs is commonly used. Mister is bad, Madam worse, but Mesdames and Messieurs, set among English words, are utterly execrable. Molasses. The older form melasses was more correct, being a regular plural of the Fr. melasse, from the Portu- guese melac.0. The word is a true plural, and was formerly so used : " The molasses will find their own outlets." BECKFORD'S "Account of Jamaica," 1790, vol. ii., p. 79. See also Encyclopedia Britannica, Art. " Sugar." It is now very commonly treated as a singular. News and tidings, plurals in form construed as singulars. Odds, meaning odd things, came into use about A.D. 1 500. It soon took the sense of difference or inequality, and in a hundred years began to be used as a singular : " I cannot speake Any beginning to this peevish oddes" " Othello," ii., iii., 185. and is now treated indifferently as singular or plural. Penny has two plurals. Six coins of that denomination are six pennies ; their value is sixpence, which admits of a plural of the second order, in a handful of silver sixpences. Pox, the only plural in x, standing tor pocks, the plural of pock, meaning the disease now commonly called small-pox. " Yes, I have known a lady sick of the small pocks, only to keep her from pit-holes, take cold, strike them in again, kick up her heels, and vanish." BEAUM. & FLETCH., " Fair Maid of the Inn." That pock is the true singular appears from its use as an adjective in pock-mark, pock-pitted, etc. ; yet pox is now used entirely as a singular. 272 English Grammar. Pulse. There are two quite different words. The one now in common use is from the Lat. pulsus, a stroke, a beat- ing, a throbbing. Curiously enough, it was often mistaken for a plural about the beginning of th'e present century, when doctors were wont to say : " Your pulse are weak to-day." The other word is rarely met with, but is an old collective term for peas, or, if any one prefer, pease, singular in form but construed as a plural. " And Barzillai the Gileadite of Rogelim brought beds, and basins, and earthen vessels, and wheat, and barley, and flour, and parched corn, and beans, and lentiles, and parched pulse" 2 SAM. xvii., 28. Riches Fr. richesse, richness, wealth an original singu- lar, converted into a plural through misapprehension as early as Chaucer's time. Sixpence see penny. Summons from Old Fr. semonse, a warning, is, and always was, a singular, although it has the appearance of a plural. Twelvemonth, a curious singular form for the plural twelve months. Like it but more reduced is fortnight for fourteen nights. Wages, from the singular wage, payment, used mostly in the plural form since the beginning of the fourteenth century. " The wages of labor, however, are much higher in North America than in any part of England." SMITH'S " Wealth of Nations." The old singular is often seen in these days in compound words in news-articles relating to wage-workers. After the plural forms have been ascertained, many ques- tions arise as to whether certain words are to be treated as singular or plural. If the form alone were conclusive there could be no question ; but singular nouns are often treated as plurals, and plural nouns as singular. The distinction depends in part on the form of the word, and partly on the Nouns. 273 nature of the thing or aggregate signified ; and we need not expect strict consistency in adjusting conflicting claims. Cattle, singular in form, is always treated as plural, while game, meaning wild animals hunted for food or amusement, however numerous, is singular. So is stock, which is nearly equivalent to cattle. Poultry, birds domesticated for eco- nomic reasons, and craft, a marine term, are either singular or plural. Nouns singular in form, denoting collective bodies of per- sons, are sometimes treated as singular, and at other times as plural. The only rational principle of distinction applica- ble here is, that when the aggregate acts as a unified body it is singular ; where the action or passion is individual the whole is to be treated as plural. This may be made clearer by a few examples, and first by three that I deem incorrect. " The prisoners taken in this action had their right foot cut off." HUME : " Hist, of England," chap. iv. The amputation could apply to the prisoners only as indi- viduals, and they must have had more than one right foot among them. " The circle of men was talking indiscriminately to both." " Robert Elsmere." Better, were talking. " There was a score of candles sparkling round the mantle- piece."" Vanity Fair." It was the individual candles that sparkled. We may properly say that a mob crosses a bridge, advances upon the city, fills and obstructs the streets ; but the mob shout, throw stones, and break into stores, for these are the acts of individuals, those of a collective mass. So a political party is singular in favoring or opposing a public measure, but plural in voting. We may say that an army marches at daybreak, and encamps on the bank of a river ; but we never say that it eats breakfast, puts on its shoes, or washes its face. Again a class of persons is generally plural ; an organized body, acting as such, is singular. 18 274 English Grammar. The clergy are exempt from military duty. The Senate is in session. Branches of science whose names end in -ics acoustics, hydrostatics, mathematics, optics, therapeutics are treated as singular. " Physics regulates more completely our social life than does his acquaintance with surrounding bodies regulate that of the savage." HERBERT SPENCER. A proper name, plural in form, is correctly used as a singular. " The Three Sisters (name of a brig) was spoken off Cape Hatteras." " ' The Hundred Wives ' is to be acted to-night." " ' English Bards and Scotch Reviewers ' is more bitter than witty." The national designation of this country, covering, as it now does, forty-four commonwealths, is sometimes used as a singular, and sometimes as a plural. Unquestionably it was regarded at first as plural. It is plural in the Constitu- tion, in President Washington's proclamation of April 22, 1793, and in the letters of " Pacificus " and " Helvidius " (Hamilton and Madison) ; but lately it is beginning to be used as a singular. The name is treated as a singular in the treaty with Corea, ratified May 22, 1882 ; in Elaine's reply to Gladstone in the North American Review for January, 1890, and in the article " United States," written for the Encyclopedia Britannica by Prof. Johnston. The change thus begun is as much political as grammatical. In the early days of the Republic the plurality of origin was kept more before men's minds than the unity of result. They emphasized the pluribus of the common motto rather than the unum ; but since 1865 there has been a greater feeling of nationality. Numbers expressing value, magnitude, distance, etc., and not individual entities, are properly treated as singular; yet usage, which is much divided, inclines to plurality. It will be readily seen how harsh and strained the following ex- amples are as plurals. Nouns. 275 Two hours are not long to wait. Forty degrees below zero are extremely cold. Seven feet are a great height for a man. Ten dollars are too much for these boots. The best writers very generally speak of a sum of money as a singular unit. " 6oo,oool, which was enough to procure a peace." BP. BURNET. " A thousand a year was thought a large revenue for a barris- ter." MACAULAY : " Hist.," chap. iii. " Moreover this forty millions does not * * * represent the whole amount to be expended under the Government bill." ARTHUR J. BALFOUR. But how little this is adhered to will be seen by three in- stances Sumner's " History of American Currency," and the reports of the Secretary of the Treasury and the Direc- tor of the Mint for 1889. The first treats a sum of money, stated in figures or words, as plural ; in the second it is seven- teen times plural and six times singular, in the last seventeen times plural and five times singular. The distinction does not seem to depend upon any principle. A case belongs here that involves the question of what is a plurality. Authorities usually give two definitions as if they were synonymous, which they are not, viz., " more than one," and " two or more." Now i comes under the first, but not under the second. Is it then singular or plural ? The short-hand style of trade adopts \\ cents, which is not according to sound analogy. One and a half loaves is equiv- alent to a whole and a half loaves, which certainly would not be good. The correct expression is a loaf and a half = a whole loaf and a half loaf, which would then be plural. THE CASES OF NOUNS. Number and gender are inherent and permanent. So long as they live, three men never become more or less than three ; nor do they ever change their gender and become women. But their relations to other persons and things may 276 English Grammar. change at any time, and to an indefinite extent. If I say, " The man is riding a bay horse," the man is the principal thing spoken of, and he is represented as doing something. But if I say, " The mans horse has run away," the horse becomes the chief actor, and the man is named only on account of his relation of ownership to the animal. Rela- tions between things may be expressed in several ways, of which four are quite common. First, as in this instance, a change in or addition to the word. Second, by little words whose office is to express the relations of one thing to others. Such are, of, to, by, from, with, which, from being very often placed before nouns, are called prepositions that is, placings-before. Third. Another class of relations are expressed by such words as my, your, his, etc. Fourth. Certain relations are expressed by words repre- senting action of some kind. In " The dog chased a wolf," and " A wolf chased the dog," the relation of pursuer and pursued is reversed. The first and third of these modes usually express but a small number of relations ; the fourth is limited to the one general relation of the actor and the thing acted upon, un- less we take into account the meaning of every separate verb. The second is co-extensive with the number of prepo- sitions. Although there is not entire agreement as to the definition of case, it is very generally limited to the first above model. I shall use the term to denote a modification in the form of a noun or pronoun to express a relation, not anything inher- ent in the subject of the noun or pronoun. The lively fancy of the Greeks represented that form of the noun which denoted the doer as standing upright, and all the others as falling away from it at varying angles of declination or " declension," These slanting forms they called ptoseis, or fallings. It was a foolish whim, as ground- less as it was useless. In the Greek word for a woman the essential part \s gunaik, but that form which represents her Nouns. 277 as being or doing anything is gune, a wider departure than the most prostrate of the fallen cases. The Romans trans- lated literally the ptoseis of the Greeks, and called them casus, fallings, our modern cases ; but they applied the name equally to what the latter supposed to stand upright. They called it the casus rectus, upright case, and the others casus obliqui, oblique or slanting. To their practical minds the terms denoted merely certain variations in the forms of words. There have been persons capable of arguing that what stood upright could not be falling, and that therefore what Cicero called the casus rectus was not a case at all. I am not sure that this race of subtle dialecticians is extinct yet. We come next to the question, how many cases we should reckon. To answer this the definition above given requires to be further guarded. If it should so happen that any word had two case forms that might be used interchange- ably throughout, having no difference of signification, we might properly say that there was but one case variously expressed. Again, if a certain noun had but one form to express two relations, while other nouns in the language had two forms, we should conclude that the exceptional word had two of its cases alike. It is difficult to find apt examples in English, although they are abundant in other languages, but the following will give an approximate idea. " Moses' law " and " Moses's law " are the same thing, while " sheep's wool " may mean the wool of one sheep or of several. Moreover, if the same case form be used through- out the language for three different relations, that will not make three cases, but only one ; and it matters not that its place is taken by three in some other language. What is called the ablative case in Latin expresses the place where anything is, the instrumentality with which a thing is done, and the source from which anything is obtained ; still it is only one case, and is uninfluenced by the circumstance that in Sanskrit there are three separate cases for these purposes. Hence the absurdity, so long persisted in, of assigning to English nouns precisely the cases claimed for Latin. If an 278 English Grammar. algebraic expression may be permitted, to make n cases there must be n forms with at least n corresponding func- tions. It will not matter that in some particular words a part of what are supposed to have been once separate forms have become indistinguishable. If they be preserved in some other words of the same class, they are to be recognized. In the following examples " The man is waiting, A dog bit the man, Man ! wait a moment, That is the track of a man, I gave the letter to the man, The horse was stopped by the man, Go with the man, It was taken from the man," the word man remains unchanged, and any noun in the language would yield a like result. How many cases then are here exhibited ? I think there is only one, and as it is used in so many different relations, I shall call it the common case. The term common has often been used for an an- alogous purpose in reference to the gender of mice, spar- rows, and the like, not easily distinguishable. There is one other form, that seen in the sentence, " This is the mans house," and as this always expresses possession or ownership, I shall call it by the usual name of the possessive case. Although most grammarians have recognized at least three distinctions, there is no novelty in thus limiting them to two. The same thing has been done by Ben Jonson, Charles Butler (1633), Fowle ("True English Grammar"), Web- ber, Jamieson (Rhetoric), Priestley, Ash, Bicknell, Dalton, Hyde, Clarke (London, 1853), Webster (" Imperial Grammar," 1831), Latham, Maetzner, and, at one time, by Lindley Murray. For reasons similar to those exhibited in the chapter on Word-Making, it is now generally held that case endings were originally separate words pronouns, prepositions, etc., that they became by frequent repetition closely associated with nouns, that by rapid and careless utterance they gradually lost part of their articulate sounds and ceased to be recognizable as separate words. Nay, they became in time so far reduced that often they were not distinguishable from each other, and in many instances not a vestige of them was left. But at their best, as known to us, they have Nouns. 2 79 generally failed to indicate all the relations required to be expressed. The Hebrew language has eleven terminations for nouns, which I think might, without overstraining the term, be called case endings, and Magyar has twenty-four ; but both have recourse to separate words to express many relations. Latin has six cases, but employs besides forty- three prepositions. Often the prepositions render the case endings unnecessary. In languages as we now know them, separate words can often do all the work ; case endings never can. But where the latter are competent, they are the .neatest, " The parson's house " is in that respect better than " The house of the parson," two syllables shorter. Cases may thus become matter of taste rather than necessity of ornament than use. They will be prized by the scholar who cultivates elegance, not by the illiterate who need great plainness. So there is a constant tendency to lose case end- ings. The illustration (p. 280) of their gradual decrease within the Aryan family of languages will be of interest. The Sanskrit alone appears to have all the cases complete ; yet that apparent completeness is deceptive. But few words have separate forms for all the cases in the singular, and none have them in the dual and plural. Moreover, if it were possible to go back a thousand years farther into the dawn of time, we might find that the extant Sanskrit had lost a number of cases before it came within the range of vision. No noun in any of these languages, except in the almost un- inflected English and French classes, has all the forms to which it is theoretically entitled. One case, the vocative, is especially defective. When it differs from the nominative it is shorter adapted to shouts and exclamations. It is never found but in the singular. In Latin it is confined to the singulars of a not very large class of words ; and in Russian to a few archaic terms of the Church service. We see here how the number of cases has dwindled till Italian, Spanish, and French nouns have no trace of them. In English the accusative or objective case is found only in the seven mono- syllables me, thee, him, her, us, them, and whom ; and in order to follow the analogy of these pronouns a majority of 280 English Grammar. . 1 sf 'rt M H w 6 u w ^en CJ M >-) o E U en O > > > > > S o 3 rt 'rt a r^ . *^i ^ r^i ^ ^ ^ 3 [3 rG J3 O O ,C U o id IS (J M rt cj rt 3 13 *^3 H S S S S S 1 en 9 r D 2 aj rt nS s a , z/^r / Eng. ure, ur, hure, our, ourn, ours. Plur. Obj. Sansk. as-md-n, nas ; Gr. /i/ief, 7/urf; Lat. an eloquent advocate." MACLIN. " personal identity can by us be placed in nothing but conscious- ness (which is that alone which makes what we call self)" LOCKE : " Human Understanding." " Agis who saw Even Sparta's self to servile avarice sunk." THOMPSON. " The ministers for the purpose hurried thence Me, and thy crying selfe." SHAKESPEARE : " Tempest." Self is an adjective when it precedes a noun. " In all service and execution, he showed the self boldness and courage that Hannibal did." NORTH'S " Plutarch." " Love did us both with one self arrow strike." It is an adverb when it precedes an adjective. " The j^" same authors do affirm." HOLLAND'S " Plinie." " In the .y^same day entered Noah, and Shem, and Ham, and Japheth." GEN. vii., 13. 326 English Grammar. It is a pronoun only in combination with pronouns. The set/was long written separately from the personal pronoun, and began to be joined to it in the fourteenth century. During the five hundred years when English underwent its greatest changes these compound pronouns took a great variety of forms, among which were : me sylf, my selve, us seolf, us self, ourself, ouszelves, you self, he seolf, hemself, himselven, their self es, their selves, them selfe, themselfe. Out of this confusion emerged a compromise like most com- promises inconsistent, but for the present fairly well estab- lished. SINGULAR PLURAL ist person, myself ourselves 2d person, thyself, yourself yourselves ( himself \ 3d person, < herself V themselves ( itself ) Five of these forms, being those of the ist and 2d persons, unite a possessive pronoun with self, as with a noun. Himself and themselves attach the second part to what is now called the objective, formerly a dative case. This may be due in part to an imitation of French expressions like lui m$me. Her- self and itself are ambiguous, but may be considered as formed on the model of himself. The prevailing construc- tion then of self is as a noun preceded by an adjective or a possessive. The two exceptions would be overcome by rendering the expressions still more emphatic. We might say : his own great self, not him own great self ; their wise selves, not them wise selves. The admission of a plural makes self 'a noun and not an adjective. We also admit of whose self and ones self. No- body ventures on whomself ; and although oneself is met with, it is against the analogy and prevailing usage of the language. Kings and their representatives alone employ the peculiar form ourself. Pronouns. 327 There is no separate objective case, but a possessive is formed by substituting own for self or selves, in which case the pronominal part is always possessive my own, his own, their own. This own, is a past participle of owe, 1 to possess, for which we now use exclusively the strengthened form own. THE INTERROGATIVE PRONOUN. The root of the interrogative pronoun is in Sanskrit ka, which becomes in Greek HO and no ; Latin, quo ; Goth, and A.-S., kva; Eng., wha and who. The distinction of gender began to fail in Gothic ; Saxon and English do not distin- guish between masculine and feminine. T is throughout a sign of the neuter. The A.-S. declension was : MASC. AND FEM. NEUTER. NOM. hwa hwset GEN. hwaes hwaes DAT. hwdm hwam Accus. hwone hwaet INSTR. hwy hwy Of these forms we have now entirely dropped the Accus. hwone, and all the others have been more or less changed or disguised. We place the w before, but still pronounce it after the h. For hwd we write who and say hu, although I have many a time heard it pronounced hwo ; and in the straths and glens of Scotland it is still called hwa. For hwas we write whose, disguising its possessive character by adding e after s, and we call it hooz. As usual with pro- nouns, we substitute the dative for the accusative, write whom and pronounce it hoom. These are never used as adjectives with nouns, and they apply only to persons. What from kwcet without a noun retains its neuter char- acter, but used adjectively is applicable to either persons or things. " What light is that ? " " or what king, going to war against another king," etc. 1 The sense of indebtedness comes from the possession of a thing not paid for. 328 English Grammar. The instrumental hwy, by what, or by what means, becomes the adverb why. From the interrogative we have two derivative pronouns, whether and which, also the adverbs how, when, where, whither, whence. Whether, having the old dual termination er, is which of the two. Which was originally formed by adding He, like, to the instrumental hwy, or hwi, and was thus equivalent to the Latin qualis, of what kind. It be- came in A.-S. hwilc or hwelc ; and in the course of its trans- formation h and w changed places, / was dropped out, and c became ch. The ancient word may perhaps still be heard in Scotland. " Whilk cause is the best I cannot say." SCOTT. Which has but one form for all occasions, is either singu- lar or plural, and is used with or without a noun. While who implies no knowledge of the person inquired about, which refers to the undetermined member of an aggregate known collectively. " and he said, I have an errand to thee, O captain. And Jehu said, Unto which of all of us ? And he said, To thee, O captain." 2 KINGS ix., 5. RELATIVE PRONOUNS. Interrogative pronouns may pass imperceptibly into a usage in which there is no trace of an interrogation. In conversing about a picture, for example, the very same words in the same order may occur in several sentences, shading off from a distinct question to no question at all : Who painted it } I wish to know who painted it. I do not know -who painted it. Here is the son of the artist -who painted it. In the last example who is said to be a relative pronoun and to relate to the preceding noun, artist, which is called the antecedent that which goes before. Who, what, which Pronouns. 329 were interrogatives before they were relatives, and began to assume the latter character in the twelfth century. There were, however, other and older relatives. The nature of a relative pronoun can be made clearer by a few examples than by any definition : 1. " There was a man in our town And he was wondrous wise." Here are two distinct statements connected by and, two subjects, man and he, and the verb was repeated. 2. " There was a man in our town Who was wondrous wise." In this instance, as in most, who = and he. 3. " There was a wondrous wise man in our town." Only one proposition, subject, and verb. Many relative clauses, but not all, can be put into the first form, but if we had sufficient adjective power, all could be expressed in the third. " He was the only candidate who was accepted," is not the same as " He was the only candidate, and he was accepted." It is equivalent to " He was the only candidate accepted" Again : " There was no one there who could swim," cannot be put into the form : " There was no one there, and he could swim " ; but we might say : " There was no one there able to swim." But relative clauses, put into the form of adjectives, would sometimes be lengthy and cumbrous, and placed after the noun, and so contrary to the genius of the language. The 330 English Grammar. Germans have a habit of making up long adjective phrases, which may be illustrated by a close translation of Apart of a sentence from Friedrich Schlegel, combining the parts of each phrase by hyphens : " The harmonious majesty of the opposite-standing Magdalen, whose consummate beauty in the toward-thc-beholder-turned feat- ures, is strikingly like to the Dresden Madonna, reminds us of the sweet harmony of the in-eternal-beatilude-blessed spirits, which in the magic tones of earthly music," and so on to the end of the sentence, that is, of the page. Now if we had this habit, we might dispense at will with relative pronouns, although it is not clear that our style would be thereby improved. Thus instead of saying : " There are still traces of the canal which was cut by Xerxes across Mount Athos," we could say : " There still are traces of the by-Xerxes-across-Mount-Athos-eut canal." Relative pronouns save us from such lumbering expressions, and, with the help of a second verb, dexterously turn a sen- tence into a new direption, from which it may be again deflected into a third. What and which as relatives differ a little from the same words as interrogatives. By a kind of poetic license which takes the possessive whose, " that sweet bird whose music was a storm." SHELLEY. " The sunken glen whose sunless shrubs must weep." BYRON. What has the peculiarity of being both antecedent and rela- tive equal to that which. Which and what are sometimes used with nouns, but oftenest without. " What books he wished, he read : What sage to hear, he heard ; what scenes to see, He saw." POLLOCK'S " Course of Time." Pronouns. 331 The quaint old phrase, what time, meaning at that time when, is found only in poetry : " What time the mighty moon was gathering light, Love paced the thymy plots of Paradise." TENNYSON. Which was often preceded by the : " In the which ye also walked sometime." COL. in., 7. More curious was the practice of requiring the three words the which that to make a single relative pronoun. " And the monstre answerde him, and seyde, he was a dedly creature, suche as God hadde formed, and dwelled in the des- ertes, in purchasynge his sustynance ; and besoughte the here- myte, that he wolde preye God for him, the whiche that cam from hevene for to saven all mankynde, and was born of a mayden, and suffred passioun and dethe (as we well knowen)." Sir JOHN MANDEVILLE. It has been already said that the interrogatives were not employed as relatives until the twelfth century. What and which came first, and Dr. Morris assures us that who was not so used until the fourteenth, nor common before the six- teenth. This priority of which explains its occurrence in old books where we should now use who. There are probably persons still living who can remember the first sentence of the Lord's Prayer neither an error nor an exception. "But Paul said, I am a man which am a Jew of Tarsus." ACTS xxi., 39. The older relatives were : first, an indeclinable the ; second, the pronoun of the third person distinguished (page 329) as No. 2, se, seo, thaet ; third, these latter combined with the ; fourth, swa, equal to so or as. In the Saxon Chronicle King Henry acts "be thaere rede the him abuton waeron." (by their rede that were about him.) " Man waes fram Gode asend thces nama waes Johannes." (There was a man sent from God whose name was John.) JOHN i., 6. 33 2 English Grammar. " Ne geseah naefre nan man God, buton se ancenneda Sunu hit cythde, se ys on hys Faeder bearme." (Not saw never no man God, unless the only begotten Son manifested it, who is in his Father's bosom.) JOHN i., 18. " He toe the recless & te blod, & 3ede upp to thatt allterr, Thatt was withthinnenn wa3herifft." (He took the incense and the blood, And went up to that altar, That was within the veil.) "Ormulum," 1200. In the "Lives of the Saints" (1295) occurs, " thulke hous as he was inne ibore," (the same house which he was born in) ; and Bishop Bonner (1538) complains of Hooper for mis- quoting him making him say : " the same as was hanged," instead of, "the same that was hanged." As was then beginning to go out of fashion. Still it sometimes found a place in literature as late as the age of Queen Anne. " he marches up and attacks their main body, but are opposed again by a party of men as lay,".etc. Tatler, 1709. Steele, in the Spectator, with the ignorance of English philology so common in that age, presents the " Humble Petition of Who and Which against the upstart Jack Sprat, That, now trying to supplant them." The truth was, they were supplanting That. Perhaps he was not acquainted with the English Psalter of 1380. " Blesse thou, my soule, to the Lord ! and wile thou not forSete all the Seldingus of him. That hath mercy to alle thi wickednessis ; that helith alle thin infirmyties. That a3en-bieth fro deth thi lif ; that crowneth thee in mercy and mercy-doingis. That fulfilleth in goode thingus thy deseyr." Pronouns. 333 In all ages of the English tongue that has been the standard relative of the body of the people, and to this day which is stiff and formal, suggestive of the student's lamp or the pedagogue's birch. Here is an excellent example : " This is the cock that crew in the morn, Unto the farmer sowing his corn, That met the priest with his pen and ink-horn, That married the man so tattered and torn, That kissed the maiden all forlorn, That milked the cow with the crumpled horn, That tossed the dog, that worried the cat, That killed the rat, that ate the malt, Ttiatlzy in the house, that Jack built." This familiar word occurs here eleven times ; and to replace it by which and zvho would destroy the rippling rhythm that has delighted the young ears of so many generations. ADJECTIVE PRONOUNS. If all be included that may with propriety be placed in this class, it becomes a very large one. Its limits are as usual rather arbitrary ; for its members are not used exclu- sively as adjectives, nor are they the only ones so employed. Several of those already treated of are often associated with nouns in the manner of adjectives. Indeed, who is the only pronoun never associated with a noun. The pronouns of the first and second persons are never adjectives. When followed by nouns, the former is the leading word to which the latter is added. It is not the pronoun that is set to help . out the noun. But the words of the present class are habitually, and some of them exclusively, employed with nouns. I shall include two groups not very generally treated as adjective pronouns, and shall make five sub- divisions. I. What are commonly called demonstrative pronouns, as if the speaker pointed with the finger at the thing spoken of. They are three in number this, with its plural these; that, with its plural those; and yon w yonder. The first two 334 English Grammar. have been spoken of under the head of the pronoun of the third person. Strictly, this refers to what is near the speaker ; that, to what is near the hearer ; and yon to something remote from both. The words are applied also to ideas, opinions, actions, sayings, etc. When two things have just been named in contrast, the first is referred to as that, the last as this. " What conscience dictates to be done, Or warns me not to do, This, teach me more than hell to shun, That, more than heaven pursue." POPE. Often we refer to something we are going to say as this, and to what we have just said as that. " Never break a bridge you yourself -may have to cross : re- member that" " Store this among your treasures of wisdom : If you cannot keep your own secret, do not expect another to do it." Like the Latin iste, this is sometimes employed to impart a shade of contempt : " There will be no end of such fantastical writers as this Mr. Harris, who takes fustian for philosophy." TOOKE'S " Diversions of Purley." " this kind, this due degree Of blindness, weakness, heaven bestows on thee." POPE : " Essay on Man," i., 233. On the other hand that is used to magnify : " When languishing with love-sick eyes, That great, that charming man you see." ADDISON : " Rosamond," ii., 6. " There Charles confronted the High Court of Justice with that placid courage which has half redeemed his fame." MACAULAY : " Warren Hastings." Yon is now out of use in prose, and rare in poetry. The later form yonder is frequent in prose as an adverb of place : Pronoims. 335 "And they too have a voice, yon piles of snow." COLERIDGE'S " Hymn." " Near yonder copse, where once a garden smiled." GOLDSMITH'S " Deserted Village." Like many other old genuine words, yon is more common in Scotland : " There 's auld Rob Morris that wins vs\yon glen, He 's the king o' guid fellows, the wale o' auld men." BURNS. II. Two little words that have not much in common are usually made into a class by themselves and called Articles. They are an, abbreviated to a before a consonant sound, and the, to which Dr. Latham would add no. An is the Saxon numeral dn, one, fairly well preserved, and, as remarked, sometimes drops the n. " I have caught a fox," and " I have caught one fox," state precisely the same fact ; but the latter carries an implication that there was some thought of catching more than one, and thus lays stress on the number. It is very clear then why an or a precedes only a singular noun. Down to the close of the last century the final n was com- monly retained before the sound of h ; and we read of an house, an high day, an hen, an hog, an heap. At present the practice is to drop it before all consonant sounds, including h,y, and w ; excepting, however, words beginning with h and accented on the second syllable. In such cases the h is scarcely heard. The adjective pronoun no is merely the negative of the preceding. One of the forms taken by the first numeral in the middle ages was o : " Anon he let two cofres make, Of o semblance and o make." GOWER. Adding to this the n of the negative particle ne, makes the compound no. A fuller form is none, but now slightly differ- ent grammatically. Formerly none, like an, was used before a vowel. 336 English Grammar. " He maketh the devices of the people of none effect." PSALM xxxiii., 10. but such a distinction is no longer observed. The is from the same source as that. Our Saxon ances- tors occasionally employed an indeclinable form the, but oftener the pronoun se, sea, thcet, varied in gender, number, and case to suit the noun to which it was applied. An, or a, no, and the are thus weakened substitutes for one, no one, and that or those. An does not indicate any particular one, while the relates to something so well understood that its identity is not likely to be mistaken. " I hear a dog barking "; " I hear the dog barking." The first may be any dog ; the second is a particular animal well understood between us. " There was a man with a monkey here yesterday." " There is the man with the monkey again." Having once seen and spoken of him he is now familiar. III. My, thy, her, our, your, their, are called possessive pro- nouns, as distinguished from the possessive cases of the per- sonal pronouns. They are of course originally the same. My and thy are shortened forms of mine and thine, in the same manner as a is a clipped an. The Bible says " mine house," " thine head." The possessive pronouns that end in r are original, and their former places have been taken by forms with an added s. fits remains the same for both pur- poses, as it does not readily admit of a second s. Its is now almost exclusively used as an adjective pronoun. Speaking of a bird we might say : " That is its nest " ; we should scarcely ever say : " That nest is its " ; yet that is the way in which Shakespeare used the word : " Each following day Became the next dayes master, till the last Made former Wonders it's" " Henry VIII.," i, x. Pronouns. 337 The words of this and the preceding subdivision are used only before nouns. We may say, " This is our house," but not, " This house is our' IV. Numerals. Words denoting number are commonly classed among adjectives, but with a misgiving that their position is insecure, like that of the bats among the birds and beasts. They express no quality or characteristic, ad- mit of no degrees of comparison, and, like pronouns, are " names for everything." It is not necessary to give a series of numerals, as they are equally well known to everybody. They offer an ex- ample of forms remarkably well preserved in the main, yet occasionally suffering great changes, as will be seen by the table on page 338, of the first twelve in several languages of the Aryan family. The first column is a conjectural original from which all the others may be supposed to be derived. The Sanskrit and Gypsy words for one are evidently not from the source common to all the others. The s or cs, added to several of the first numerals, is an inflectional termination, and no part of the original words. The same is true of final r, or the doubled consonant, in Norse or Icelandic. The d in the Russian od-in is excrescent. Two, three, and six a multiple of two and three are the numbers best preserved. Whether or not this has any con- nection with the mystic character often attributed to the number three I cannot say. The cipher 3 is still the same in Sanskrit and in European books. Eleven and twelve are generally one-ten, two-ten ; but are constructed on a different principle in Lithuanian and the Teutonic dialects. This is most readily seen in Gothic. Bopp is no doubt correct in his conjecture that the -/if is akin to the English leave, left. The line of thought would thus be ten-and-one-left, ten-and-tzvo-left, shortened to one- left, two-left. In third, thirteen, and thirty r has changed places, as it often does, with the adjacent vowel. 338 English Grammar. jj e V O V o O a 4J M 1 1 seven .3 '3 g c V U 13 1 c O d 1 s 4> d ' bfl | e " '' 1u M-c o 3 BB c rg -0 5 'at *o I/) d .> S a C Zi V 1> I H c IH M T S 3 d 'S 1 1 2, m I 3 -rt 1 3 i 3 U i 3 '3 U) "I o IM S WJ C 3 -S rt S3 3 '5 3 'd >; c V eu z t/1 M ^ a 4) 3 g S U s g 3 5 1 1 f g IS ^ ^ trP .S O oO i S 6 O 1 novem o V u T (3 3 U 1 13 d H U M "CO 5? a 2- co ^ Q. h ^ ^ i: CO *a co a CO a X CO <0 O * o CO d -co ' I- k -w l> h k X ^ co J 3 O <0 K> r~ ij <0 "CO "O -co tco <0 d d Z Ji a 1 M T) O d 1 d jj O >, to 43 U osem > d V d -3 o > d tJO M o d 12 H d o d d "d u L^ u d u o 'o (3 4j3 '53 T O d 1 K d M %>M j3 d < 1 d (3