This question, why a full third of Shakespeare's verbal riches was never brought to light more than once, is probably one which nobody can at present answer even to his own satisfaction. Yet the phenomenon is so remarkable that every one will try after his own fashion to account for it. My own attempt at a provisional explanation I will present in the latter part of this paper.
Let us first, however, notice another question concerning the απαξ λεγόμενα—namely, that which respects their origin. Where did they come from? how far did Shakespeare make them? and how far were they ready to his hand? No approach to answering this inquiry can be made for some years. Yet as to this matter let us rejoice that the unique dictionary of the British Philological Society is now near publication. This work, slowly elaborated by thousands of co-workers in many devious walks of study on both sides of the Atlantic, aims to exhibit the first appearance in a book of every English word. In regard to the great bulk of Shakespeare's diction it will enable us ten years hence to determine how much of it was known to literature before him, and how much of it he himself gathered or gleaned in highways and byways, or caused to ramify and effloresce from Saxon or classical roots and trunks, thus "endowing his purposes with words to make them known." Meantime, we are left to conjectures. As of his own coinage I should set down such vocables as motley-minded, mirth-moving, mockable, marbled, martyred, merriness, marrowless, mightful, multipotent, masterdom, monarchize, etc. etc.
But, however much of his linguistic treasury Shakespeare shall be proved to have inherited ready-made—whatever scraps he may have stolen at the feast of languages—it is clear that he was an imperial creator of language, and lived while his mother-tongue was still plastic. Having a mint of phrases in his own brain, well might he speak with the contempt he does of those "fools who for a tricksy word defy the matter;" that is, slight or disregard it. He never needed to do that. Words were "correspondent to his command, and, Ariel-like, did his spiriting gently."
In a thousand cases, however, Shakespeare cannot have rejected words through fear lest he should repeat them. It has taken three centuries for the world to ferret out his απαξ λεγόμενα: can we believe that he knew them all himself? Unless he were the Providence which numbers all hairs of the head, he had not got the start of the majestic world so far as that, however myriad-minded we may consider him. An instinct which would have rendered him aware of each and every individual of five thousand that he had employed once only would be as inconceivable as that of Falstaff, which made him discern the heir-apparent in Prince Hal when disguised as a highwayman. In short, Shakespeare could not be conscious of all the words he had once used, more than Brigham Young could recognize all the wives he had once wedded.
In the absence of other theories concerning the reasons for Shakespeare's απαξ λεγόμενα being so abundant, I throw out a suggestion of my own till a better one shall supplant it.
Shakespeare's forte lay in characterization, and that endlessly diversified. But when he sketched each several character it seems that he was never content till he had either found or fabricated the aptest words possible for representing its form and pressure most true to life. No two characters being identical in any particular more than two faces are, no two descriptions, as drawn by his genius, could repeat many of the selfsame characterizing words. Each of his vocables thus became like each of the seven thousand constituents of a locomotive, which fits the one niche it was ordained to fill, but everywhere else is out of place, and even dislocated. The more numerous his ethical differentiations, the more his language was differentiated.
His personages were as multifarious as have been portrayed by the whole band of Italian painters; but, as a wizard in words, he resembled the magician in mosaic, who can delineate in stone every feature of those portraits because he can discriminate and imitate shades of color more numberless than even Shakespeare's words.
It is hard to believe that the Shakespearian characters were born, like Athene from the brain of Zeus, in panoplied perfection. They grew. The play of Troilus was a dozen years in growth. According to the best commentators, "Shakespeare, after having sketched out a play on the fashion of his youthful taste and skill, returned in after years to enlarge it, remodel it, and enrich it with the matured fruits of years of observation and reflection. Love's Labor Lost first appeared in print with the annunciation that it was 'newly corrected and augmented,' and Cymbeline was an entire rifacimento of an early dramatic attempt, showing not only matured fulness of thought, but laboring intensity of compressed expression." So speaks Verplanck, and his utterance is endorsed by Richard Grant White.
Such being the facts, it is clear that Shakespeare treated his dramas as Guido did the Cleopatra, which he would not let leave his studio till ten years after the non-artistic world deemed that portrait fully finished. Meantime, the painter in moments of inspiration was pencilling his canvas with curious touches, each approximating nearer his ideal. So the poet sought to find out acceptable words, or what he terms "an army of good words." He poured his new wine into new bottles, and never was at rest till he had arrayed his ideas in that fitness of phrase which comes only by fits.
Had he survived fifty years longer, I suppose he would to the last have been perfecting his phrases, as we read in Dionysius of Halicarnassus that Plato up to the age of eighty-one was "combing and curling, and weaving and unweaving, his writings after a variety of fashions." Possibly, the great dramatist would at last have corrected one of his couplets as a modern commentator has done for him, so that it would stand,