Varro: DE LINGVA LATINA

*** Insignis eorum est error qui malunt quae nesciunt docere quam discere quae ignorant: in quo fuit Crates, nobilis grammaticus, qui fretus Chrysippo, homine acutissimo qui reliquit peri anomalias III libros, contra analogian atque Aristarchum est nixus, sed ita, ut scripta indicant eius, ut neutrius videatur pervidisse voluntatem, quod et Chrysippus de inaequabilitate cum scribit sermonis, propositum habet ostendere similes res dissimilibus verbis et dissimiles similibus esse vocabulis notatas, id quod est verum, et quod Aristarchus, de aequabilitate cum scribit eiusdem,vuerborum similitudinem quandam in inclinatione sequi iubet, quoad patiatur conseutudo.

*** Distinguished is the error of those who prefer to teach what they do not know rather than to learn what they are ignorant of: in this was Crates, a noble grammarian, who, relying on Chrysippus, a most sharp man who left 3 books peri anomalias (“On Anomalies”), took his stand against analogy and Aristarchus; but in such a way, as his writings indicate, that he seems to have perceived the intention of neither, because both Chrysippus, when he writes about the inequality of speech, has the purpose to show that similar things are marked by dissimilar words and dissimilar by similar vocables—which is true—and Aristarchus, when he writes about the equality of the same, bids that a certain similarity of words be followed in inclination, so far as custom allows.

Sed ii qui in loquendo partim sequi iubent nos consuetudinem partim rationem, non tam discrepant, quod consuetudo et analogia coniunctiores sunt inter se quam iei credunt, quod est nata ex quadam consuetudine analogia et ex hac consuetudine item anomalia. Quare quod consuetudo ex dissimilibus et similibus verbis eorumque declinationibus constat, neque anomalia neque analogia est repudianda, nisi si non est homo ex anima, quod ex corpore et anima.

But those who in speaking bid us partly to follow custom and partly reason do not differ so much, because custom and analogy are more closely conjoined with each other than they think, since analogy was born from a certain custom, and from this same custom likewise anomaly. Wherefore, since custom consists of dissimilar and similar words and their declensions, neither anomaly nor analogy is to be repudiated, unless man is not from soul—for he is from body and soul.

Sed ea quae dicam quo facilius pervideri possint, prius de trinis copulis discernendum (nam confusim ex utraque parte pleraque dicuntur, quorum alia ad aliam referri debent summam): primum de copulis naturae et usuis: haec enim duo sunt quo deriguntur diversa, quod aliud est dicere esse verborum analogias, aliud dicere uti oportere analogiis; secundum de copulis multitudinis ac finis, utrum omnium verborum dicatur esse analogiarum usus an maioris partis; tertium de copulis personarum, qui eis debeant uti, quae sunt plures.

But, in order that the things I shall say may be more easily seen through, it must first be distinguished according to three couplings (for on both sides most matters are spoken confusedly, of which some ought to be referred to one sum, others to another): first, about the couplings of nature and of use; for these two are that by which the diverse are directed, since it is one thing to say that there are analogies of words, another to say that one ought to use analogies; second, about the couplings of multitude and of end, whether the use of analogies is said to be of all words or of the greater part; third, about the couplings of persons, who ought to use them, which are several.

Alia enim populi universi, alia singulorum, et de ieis non eadem oratoris et poetae, quod eorum non idem ius. Itaque populus universus debet in omnibus verbis uti analogia et, si perperam est consuetus, corrigere se ipsum, cum orator non debeat in omnibus uti, quod sine offensione non potest facere, cum poeta transilire lineas impune possit.

For one things belong to the people as a whole, other things to individuals, and concerning these the same rules are not those of the orator and the poet, because their right is not the same. Therefore the people as a whole ought to use analogy in all words and, if it has been wrongly accustomed, to correct itself, whereas the orator ought not to use it in all things, because he cannot do this without offense, whereas the poet can leap over the lines with impunity.

Populus enim in sua potestate, singuli in illius: itaque ut suam quisque consuetudinem, si mala est, corrigere debet, sic populus suam. Ego populi consuetudinis non sum ut dominus, at ille meae est. Ut rationi optemperare debet gubernator, gubernatori unus quisque in navi, sic populus rationi, nos singuli populo.

For the people are in their own power, individuals in that of the people: therefore, as each person ought to correct his own custom, if it is bad, so the people theirs. I am not, as master, of the people’s custom, but it is, as master, of mine. As the governor (helmsman) ought to obey reason, and each person on the ship the governor, so the people reason, we individuals the people.

Quare ad quamcumque summam in dicendo referam si animadvertes, intelleges, utrum dicatur analogia esse an uti oportere ea; itemque intelleges si ad analogiam usum loquendi oporteat redigere, tum dici id in populum aliter ac in singulos nec idem de omnibus dici in eum qui sit in populo.

Therefore, if you notice to whatever chief standard in speaking I refer matters, you will understand whether it should be said that there is analogy or that one ought to use it; likewise you will understand, if the usage of speech ought to be brought back to analogy, then that this is to be said with respect to the people differently than with respect to individuals, and that the same is not to be said of everyone as of the one who is in the people.

II.

Nunc iam primum dicam pro universa analogia, cur non modo non videatur esse reprehendenda, sed etiam cur in usu quodammodo sequenda; secundo de singulis criminibus, quibus rebus possint quae dicta sunt contra solvi, dicam ita ut generatim comprehendam et ea quae in priore libro sunt dicta et ea quae possunt dici atque illic praeterii.

Now, first, I will speak on behalf of universal analogy, why it not only does not seem to be reprehensible, but also why in use it is in some measure to be followed; secondly, concerning the individual charges, by what means the things that have been said against it can be resolved, I will speak so as to comprehend in general both those things which were said in the prior book and those which can be said and which I passed over there.

III.

Primum quod aiunt, qui bene loqui velit consuetudinem sequi oportere, non rationem similitudinum, quod, alteram si neglegat, sine offensione facere non possit, alteram si sequatur, quod sine reprehensione non sit futurum, errant, quod qui in loquendo consuetudinem qua oportet uti sequitur, eam sequitur non sine ratione.

First, as to what they say, that he who wishes to speak well ought to follow custom (usage), not the rationale of similitudes (analogy), since, if he neglect the one, he cannot do it without offense, and if he follow the other, it will not be without reprehension, they err, because he who in speaking follows the custom which it is proper to employ follows it not without reason.

IV.

Nam vocabula ac verba quae declinamus similiter, ea in consuetudine esse videmus et ad eam conferimus et, si quid est erratum, non sine ea corrigimus. Nam ut, qui triclinium constrarunt, si quem lectum de tribus unum imparem posuerunt aut de paribus nimium aut parum produxerunt, una corrigimus et ad consuetudinem communem et ad aliorum tricliniorum analogias, sic si quis in oratione in pronuntiando ita declinat verba ut dicat disparia, quod peccat redigere debemus ad ceterorum similiumvuerborum rationem.

For the nouns and words which we decline similarly, we see these to be in usage and we refer them to that, and, if anything has erred, we do not correct it without that. For just as those who have furnished a triclinium, if they have set one couch out of three unequal, or from equal ones have drawn one out too far or too little, we correct it at once both according to the common custom and according to the analogies of other triclinia; so, if anyone in oration, in pronouncing, declines words so as to say disparate things, what he has done amiss we ought to bring back to the rule of the other similar words.

Cum duo peccati genera sint in declinatione, unum quod in consuetudinem perperam receptum est, alterum quod nondum est et perperam dicatur, unum dant non oportere dici, quod non sit in consuetudine, alterum non conceditur quin ita dicatur, ut sit similiter, cum id faciant, ac, si quis puerorum per delicias pedes male ponere atque imitari vatias coeperit, hos corrigi oportere si concedant, contra si quis in consuetudine ambulandi iam factus sit vatia aut conpernis, si eum corrigi non concedant.

Since there are two kinds of fault in declension, one which has been wrongly received into custom, the other which is not yet (in custom) and is said wrongly, for the one they grant it ought not to be said, because it is not in custom; for the other they do not concede that it be said so as to be similar, when they do this, just as, if any of the boys, for delight, should begin to set his feet badly and imitate the knock‑kneed, they concede that these ought to be corrected; contrariwise, if someone in the habit of walking has already become knock‑kneed or bow‑legged, they do not concede that he be corrected.

Non sequitur, ut stulte faciant qui pueris in geniculis alligent serperastra, ut eorum depravata corrigant crura? Cum vituperandus, non sit medicus qui e longinqua mala consuetudine aegrum in meliorem traducit, quare reprehendendus sit qui orationem minus valentem propter malam consuetudinem traducat in meliorem?

It does not follow that those act foolishly who bind leg-splints on boys’ knees, in order to correct their distorted legs, does it? Since a physician is not to be blamed who translates a patient from a long‑standing bad habit into a better condition, why should he be reprehended who translates speech, made less robust by bad usage, into something better?

VI.

Pictores Apelles, Protogenes, sic alii artufices egregii non reprehendundi, quod consuetudinem Miconos, Dioris, Arimmae, etiam superiorum non sunt secuti: Aristophanes improbandus, qui potius in quibusdam veritatem quam consuetudinem secutus?

Are the painters Apelles, Protogenes, and likewise other outstanding artificers not to be reprehended, because they did not follow the custom of Micon, Diores, Arimma, even of their superiors: is Aristophanes to be disapproved, who rather in certain matters followed truth rather than custom?

VII.

Quod si viri sapientissimi, et in re militari et in aliis rebus multa contra veterem consuetudinem cum essent ausi, laudati, despiciendi sunt qui potiorem dicunt oportere esse consuetudinem ratione.

But if the most sapient men, both in military affairs and in other matters, having dared many things against the old custom, were praised, those are to be despised who say that custom ought to be preferable to reason.

VIII.

An cum quis perperam consuerit quid facere in civitate, non modo non patiemur, sed etiam poena afficiemus, idem si quis perperam consuerit dicere verbum, non corrigemus, cum id fiat sine poena?

Or, when someone has wrongly become accustomed to do something in the commonwealth, not only shall we not allow it, but we shall even afflict him with punishment; shall we not likewise correct it, if someone has wrongly become accustomed to say a word, since that is done without penalty?

IX.

Et hi qui pueros in ludum mittunt, ut discant quae nesciunt verba quemadmodum scribant, idem barbatos qui ignorabunt verba quemadmodum oporteat dici non docebimus, ut sciant qua ratione conveniat dici?

And those who send boys into school, so that they may learn how to write the words they do not know—shall we not likewise teach the bearded men, who are ignorant, with respect to words, how it ought to be said, so that they may know by what rationale it is fitting to be said?

10.

Sed ut nutrix pueros a lacte non subito avellit a consuetudine, cum a cibo pristino in meliorem traducit, sic maiores in loquendo a minus commodis verbis ad ea quae sunt cum ratione modice traducere oportet. Cum sint in consuetudine contra rationem alia verba ita ut ea facile tolli possint, alia ut videantur esse fixa, quae leviter haerent ac sine offensione commutari possunt statim ad rationem corrigi oportet, quae autem sunt ita ut in praesentia corrigere nequeas quin ita dicas, his oportet, si possis, non uti: sic enim obsolescent ac postea iam obliterata facilius corrigi poterunt.

But as a nurse does not suddenly tear boys away from milk and from their custom, when she transfers them from their former food to a better, so it is proper to translate adults in speaking from less commodious words to those which are with reason, in a measured way. Since in custom there are certain words contrary to reason, some such that they can easily be removed, others such that they seem to be fixed, those which adhere lightly and can be changed without offense ought at once to be corrected to reason; but those which are such that for the present you cannot correct without having to speak so, these it is proper, if you can, not to use: for thus they will grow obsolete, and afterward, once already obliterated, they will be able to be corrected more easily.

XI.

11.

Quas novas verbi declinationes ratione introductas respuet forum, his boni poetae, maxime scaenici, consuetudine subigere aures populi debent, quod poetae multum possunt in hoc: propter eos quaedam verba in declinatione melius, quaedam deterius dicuntur. Consuetudo loquendi est in motu: itaque solent fieri et meliora deteriora et deteriora meliora; verba perperam dicta apud antiquos aliquos propter poetas non modo nunc dicuntur recte, sed etiam quae ratione dicta sunt tum, nunc perperam dicuntur.

The new declensions of a word, introduced by reason, which the forum rejects, to these the good poets, especially the stage-poets, ought by usage to train the people’s ears; for poets can do much in this: on account of them certain words in declension are said better, certain worse. The usage of speaking is in motion; and so it is wont to happen both that the better become worse and the worse better; words incorrectly said among some ancients, on account of the poets, are not only now said rightly, but even those which were then said according to reason are now said incorrectly.

XII.

12.

Quare qui ad consuetudinem nos vocant, si ad rectam, sequemur: in eo quoque enim est analogia; si ad eam invitant quae est depravata, nihilo magis sequemur, nisi cum erit necesse, quam in ceteris rebus mala exempla: nam ea quoque, cum aliqua vis urget, inviti sequemur.

Wherefore, those who call us to usage—if to the right one, we shall follow; for in that too there is analogy; but if they invite to that which is depraved, we shall follow no more than bad examples in other matters, unless it will be necessary; for those also, when some force urges, we shall follow unwillingly.

XIII.

13.

Neque enim Lysippus artificum priorum potius secutus est vitiosa quam artem; sic populus facere debet, etiam singuli, sine offensione quod fiat populi.

For neither did Lysippus follow the faulty things of earlier artisans rather than the art; thus the people ought to act, and individuals too, without offense, in what is done by the people.

Qui amissa non modo quaerant, sed etiam quod indicium dent, idem, ex sermone si quid deperiit, non modo nihil impendunt ut requirant, sed etiam contra indices repugnant ne restituatur?

Those who not only seek things lost, but even offer an informer’s reward—do these same people, if anything has perished from speech, not only expend nothing to inquire for it, but even resist informers lest it be restored?

Verbum quod novum et ratione introductum quo minus recipiamus, vitare non debemus.

We ought not to avoid a word which is new and introduced with reason, so as to be the less willing to receive it.

XIV.

14.

Nam ad usum in vestimentis aedificiis supellectili novitati non impedit vetus consuetudo: quem enim amor assuetudinis potius in pannis possessorem retinet, quem ad nova vestimenta traducit?

For in use in vestments, edifices, [and] household furniture, old custom does not impede novelty: whom, indeed, does the love of habituation keep the possessor rather in rags, whom does it lead across to new vestments?

XV.

15.

An non saepe veteres leges abrogatae novis cedunt?

Do not old laws, once abrogated, often yield to new ones?

XVI.

16.

Nonne inusitatis formis vasorum recentibus e Graecia adlatis obliteratae antiquae consuetudinis sinorum et capularum species? His formis vocabulorum incontaminatis uti nolent quas docuerit ratio propter consuetudinem veterem? Et tantum inter duos sensus interesse volunt, ut oculis semper aliquas figuras supellectilis novas conquirant, contra auris expertis velint esse?

Are not the unusual forms of vessels, recently brought in from Greece, obliterating the types of hollows and handles of ancient custom? Will they, because of the old custom, be unwilling to use for these forms the uncontaminated vocabulary which reason has taught? And do they want there to be so much difference between the two senses, that with their eyes they are always seeking out some new figures of household-ware, but, on the contrary, wish to be against experienced ears?

XVII.

17.

Quotus quisque iam servos habet priscis nominibus? Quae mulier suum instrumentum vestis atque auri veteribus vocabulis appellat? Sed indoctis non tam irascendum quam huiusce pravitatis patronis.

How many nowadays have slaves with ancient names? What woman calls her own accoutrement of clothing and of gold by old vocables? But we should be angry not so much with the unlearned as with the patrons of this very perversity.

Si enim usquequaque non esset analogia, tum sequebatur, ut in verbis quoque non esset, non, cum esset usquequaque, ut est, non esse in verbis.

For if analogy were not everywhere, then it would follow that it was not in words as well; not, when it is everywhere, as it is, that it is not in words.

XVIII.

18.

Quae enim est pars mundi quae non innumerabiles habeat analogias? Caelum an mare an terra, quae in his?

What part of the world is there that does not have innumerable analogies? The sky or the sea or the earth—what of these?

Nonne in caelo ut ab aequinoctiali circulo ad solstitialem et hinc ad septemtrionalem divisum, sic contra paribus partibus idem a bruma versum contraria parte? Non quantum polus superior abest a septemtrionali circulo et is a solstitiali, quem sol circumit cum it ad solstitium, tantundem abest inferior ab eo quem antarcticon vocant astrologi et is a brumali? Non, quemadmodum quodque signum exortum hoc anno, quotquot annis eodem modo exoritur?

Is it not in the heaven that, just as from the equinoctial circle to the solstitial, and from there to the septentrional, it is divided, so likewise on the opposite side the same measure is set off from the brumal? Is it not that, by as much as the upper pole is distant from the septentrional circle and that from the solstitial one—around which the sun circles when it goes to the solstice—by just so much the lower is distant from that which the astrologers call the Antarctic, and that one from the brumal? Is it not, just as whatever sign has risen this year, that in just so many years it rises in the same way?

Num aliter sol a bruma venit ad aequinoctium, ac contra cum ad solstitium venit, ad aequinoctialem circulum et inde ad brumam? Nonne luna, ut ab sole discedit ad aquilonem et inde redit in eandem viam, sic inde fertur ad austrum et regreditur inde? Sed quid plura de astris, ubi difficilius reperitur quid sit aut fiat in motibus dissimiliter?

Does the sun come in any other way from bruma to the equinox, and conversely, when it comes to the solstice, to the equinoctial circle and thence to bruma? Does not the moon, as it departs from the sun toward the north and from there returns into the same path, likewise from there is borne toward the south and regresses from there? But why say more about the stars, where it is more difficult to discover what is or happens in their motions in a dissimilar manner?

XIX.

19.

At in mari, credo, motus non habent similitudines geminas, qui in XXIIII horis lunaribus cotidie quater se mutant, ac cum sex horis aestus creverunt, totidem decreverunt, rursus idem, itemque ab his. An hanc analogian ad diem servant, ad mensem non item, alios motus sic item cum habeant aliis inter se convenientes? De quibus in libro quem de Aestuariis feci scripsi.

But in the sea, I suppose, the motions do not have twin similitudes, which within 24 lunar hours change themselves four times daily; and when in six hours the tides have increased, by as many they have decreased—again the same, and likewise the reverse from these. Or do they keep this analogy with the day, but not with the month, since they have other motions similarly agreeing with others among themselves? About these I have written in the book which I composed, On Estuaries.

XX.

20.

Non in terra in sationibus servata analogia? Nec cuius modi in praeterito tempore fructuum genera reddidit, similia in praesenti reddit, et cuius modi tritico iacto reddidit segetes, sic hordeo sato proportione reddidit parilis? Non, ut Europa habet flumina lacus, montis campos, sic habet Asia?

Is not an analogy preserved in the earth in sowings? And does it not render in the present similar kinds of fruits as the sort it rendered in past time; and of the sort that, when wheat was cast, it returned crops, so, when barley was sown, in proportion it returned like ones? Is it not that, as Europe has rivers, lakes, mountains, plains, so Asia has?

XXI.

21.

Non in volucribus generatim servatur analogia? Non ex aquilis aquilae atque ut ex turdis qui procreantur turdi, sic ex reliquis sui cuiusque generis?

Is not analogy preserved among the birds, by kinds? Is it not from eagles that eagles come, and as from thrushes those who are procreated are thrushes, so from the rest, each of its own kind?

XXII.

22.

An aliter hoc fit quam in aere in aqua? Non hic conchae inter se generatim innumerabili numero similes? Non pisces?

Or is this done otherwise than in air and in water? Are not the shells here, among themselves, by genera, alike in innumerable number? Are not the fishes?

An e muraena fit lupus aut merula? Non bos ad bovem collatus similis, et qui ex his progenerantur inter se vituli? Etiam ubi dissimilis fetus, ut ex asino et equa mulus, tamen ibi analogia: quod ex quocumque asino et equa nascitur id est mulus aut mula, ut ex equo et asina hinnulei.

Or does a wolf or a merle (blackbird) come from a moray-eel? Is not an ox, compared to an ox, similar, and are not those begotten from these among themselves calves? Even where the offspring is dissimilar—such as from an ass and a mare a mule—still there is an analogy: whatever is born from any ass and a mare is a mule or a she‑mule, just as from a horse and a she‑ass come hinnies.

XXIII.

23.

Non sic ex viro et muliere omnis similis partus, quod pueri et puellae? Non horum ita inter se omnia similia membra, ut separatim in suo utroque genere similitudine sint paria? Non, omnes cum sint ex anima et corpore, partes quoque horum proportione similes?

Is it not thus that from man and woman all similar offspring come—that is, boys and girls? Are not all the members of these so similar among themselves, that, taken separately, in each of the two kinds they are equal in likeness? Since all are of soul and body, are not their parts also similar in proportion?

Quid ergo cum omnes animae hominum sint diuisae in octonas partes, hae inter se non proportione similes? Quinque quibus sentimus, sexta qua cogitamus, septuma qua progeneramus, octaua qua uoces mittimus? Igitur quoniam qua loquimur uoce oratio est, hanc quoque necesse est natura habere analogias: itaque habet.

What then? Since all the souls of men are divided into eight parts, are these not similar among themselves in proportion? Five by which we perceive, a sixth by which we think, a seventh by which we progenerate, an eighth by which we emit voices? Therefore, since oration is the voice with which we speak, it too must by nature have analogies: and so it has.

XXIV.

24.

An non vides, ut Graeci habeant eam quadripertitam, unam in qua sint casus, alteram in qua tempora, tertiam in qua neutrum, quartum in qua utrumque, sic nos habere? Ecquid verba nescis ut apud illos sint alia finita, alia non, sic utraque esse apud nos?

Do you not see that the Greeks have it quadripartite: one in which there are cases, another in which there are times, a third in which there is neither, a fourth in which there is both—so we have it likewise? Do you not at all know that words among them are some finite, others not, so that in like manner we have both?

Equidem non dubito, quin animadverteris item in ea innumerabilem similitudinum numerum, ut trium temporum verbi aut trium personarum.

Indeed I do not doubt that you have likewise noticed in these an innumerable number of similitudes, such as of the three tenses of the verb or of the three persons.

XXV.

25.

Quis enim potest non una animadvertisse in omni oratione esse ut legebam lego legam sic lego legis legit, cum haec eadem dicantur alias ut singula, alias ut plura significentur? Quis est tam tardus qui illas quoque non animadverterit similitudines, quibus utimur in imperando, quibus in optando, quibus in interrogando, quibus in infectis rebus, quibus in perfectis, sic in aliis discriminibus?

For who, indeed, can have failed at once to notice that in every oration there are, as I was reading, I read, I shall read; so too I read, you read, he reads, since these same things are uttered sometimes so that single things are signified, sometimes so that more things are signified? Who is so slow as not also to have noticed those similitudes which we use in commanding, which in opting (wishing), which in interrogating, which in unfinished (imperfect) matters, which in finished (perfect) ones, and so in other discriminations?

XXVI.

26.

Quare qui negant esse rationem analogiae, non vident naturam non solum orationis, sed etiam mundi; qui autem vident et sequi negant oportere, pugnant contra naturam, non contra analogian, et pugnant volsillis, non gladio, cum pauca excepta verba ex pelago sermonis populi minus usu trita afferant, cum dicant propterea analogias non esse, similiter ut, si quis viderit mutilum bovem aut luscum hominem claudicantemque equum, neget in bovum hominum et equorum natura similitudines proportione constare.

Wherefore those who deny that there is a rationale of analogy do not see the nature not only of speech but also of the world; but those who do see and deny that it ought to be followed fight against nature, not against analogy, and they fight with forceps, not with a sword, when, a few exceptions aside, they bring forward words less worn by use from the sea of the people’s speech, when they say on that account that analogies do not exist—just as, if someone were to see a mutilated ox or a one-eyed man and a limping horse, he would deny that in the nature of oxen, men, and horses similitudes are constituted by proportion.

XXVII.

27.

Qui autem duo genera esse dicunt analogiae, unum naturale, quod ut ex satis nascuntur lentibus lentes sic ex lupino lupinum, alterum voluntarium, ut in fabrica, cum vident scaenam ut in dexteriore parte sint ostia, sic esse in sinisteriore simili ratione factam, de his duobus generibus naturalem esse analogian, ut sit in motibus caeli, voluntariam non esse, quod ut quoique fabro lubitum sit possit facere partis scaenae: sic in hominum partibus esse analogias, quod eas natura faciat, in verbis non esse, quod ea homines ad suam quisque voluntatem fingat, itaque de eisdem rebus alia verba habere Graecos, alia Syros, alia Latinos: ego declinatus verborum et voluntarios et naturalis esse puto, voluntarios quibus homines vocabula imposuerint rebus quaedam, ut ab Romulo Roma, ab Tibure Tiburtes, naturales ut ab impositis vocabulis quae inclinantur in tempora aut in casus, ut ab Romulo Romuli Romulum et ab dico dicebam dixeram.

But those who say that there are two kinds of analogy, one natural, as just as from seed beans are born from beans so from lupine the lupine, the other voluntary, as in the workshop, when they see a stage so made that on the right-hand side there are doors, thus on the left-hand side it is made by a similar rationale,—of these two kinds the analogy is natural, as it is in the motions of the heaven, not voluntary, because a craftsman, as it has pleased each one, can make the parts of the stage: so in the parts of humans there are analogies, because nature makes them; in words there are not, because men fashion them each according to his own will, and thus for the same things the Greeks have certain words, the Syrians others, the Latins others: I think the declensions of words are both voluntary and natural, voluntary in which men have imposed names upon certain things, as from Romulus Roma, from Tibur Tiburtes, natural in that from the imposed names there are those that are inflected into tenses or into cases, as from Romulus Romuli Romulum and from dico dicebam dixeram.

Itaque in voluntariis declinationibus inconstantia est, in naturalibus constantia; quae utrasque quoniam iei non debeant negare esse in oratione, quom in mundi partibus omnibus sint, et declinationes verborum innumerabiles, dicendum est esse in his analogias. Neque ideo statim ea in omnibus verbis est sequenda: nam si qua perperam declinavit verba consuetudo, ut ea aliter non possint efferri sine offensione multorum, hinc rationem verborum praetermittendam ostendit loquendi ratio.

Therefore, in voluntary declensions there is inconstancy, in natural ones constancy; since they ought not to deny that both of these are in speech, seeing that they are in all parts of the world, and the declensions of words are innumerable, one must say that there are analogies in these. Nor, for that reason, is it straightway to be followed in all words: for if custom has wrongly declined certain words, so that they cannot be otherwise expressed without the offense of many, from this the manner of speaking shows that the reason (rationale) of the words is to be passed over.

XXVIII.

28.

Quod ad universam pertinet causam, cur similitudo et sit in oratione et debeat observari et quam ad finem quoque, satis dictum. Quare quod sequitur de partibus singulis deinceps expediemus ac singula crimina quae dicunt contra analogias solvemus.

As regards the universal cause—why similitude both is in speech and ought to be observed, and to what end as well—enough has been said. Wherefore what follows we shall in turn set forth concerning the several parts, and we shall resolve the individual charges which they allege against the analogies.

In quo animadvertito natura quadruplicem esse formam, ad quam in declinando accommodari debeant verba: quod debeat subesse res quae, designetur, et ut sit ea res in usu, et ut vocis natura ea sit quae significavit, ut declinari possit, et similitudo figurae verbi ut sit ea quae ex se declinatu genus prodere certum possit.

In this observe that the natural form is fourfold, to which, in declining, words ought to be accommodated: that there should underlie a thing which is designated, and that that thing be in use, and that the nature of the voice be such as has signified, so that it can be declined, and that the likeness of the figure of the word be such as can from itself, by declension, produce a certain genus.

Quo neque a terra terrus ut dicatur postulandum est, quod natura non subest, ut in hoc alterum maris, alterum feminae debeat esse; sic neque propter usum, ut Terentius significat unum, plures Terentii, postulandum est, ut sic dicamus faba et fabae: non enim in simili usu utrumque; neque ut dicimus ab Terentius Terentium, sic postulandum ut inclinemus ab A et B, quod non omnis vox natura habet declinatus.

Wherefore neither must it be postulated that from terra one should say terrus, because nature is not present, so that in this case the one ought to be of the masculine, the other of the feminine; so neither on account of usage, since Terentius signifies one, several Terentii, must it be postulated that we should thus say faba and fabae: for the two are not in similar usage; nor, because we say from Terentius Terentium, must it be postulated that we inflect from A and B, since not every voice by nature has declensions.

Neque in forma collata quaerendum solum, quid habeat in figura simile, sed etiam nonnunquam in eo quem habeat effectum. Sic enim lana Gallicana et Apula videtur imperito similis propter speciem, cum peritus Apulam emat pluris, quod in usu firmior sit. Haec nunc strictim dicta apertiora fient infra.

Nor, in the form collated, should one seek only what it has similar in figure, but also sometimes the effect which it has. For thus Gallican and Apulian wool seems similar to the unskilled because of the appearance, whereas an expert buys the Apulian at a higher price, because in use it is more firm. These things, now said briefly, will become clearer below.

Incipiam hinc.

I shall begin from here.

XXIX.

29.

Quod rogant ex qua parte oporteat simile esse verbum, a voce an a significatione, respondemus a voce; sed tamen nonnunquam quaerimus genere similiane sint quae significantur ac nomen virile cum virili conferimus, feminae cum muliebri: non quod id quod significant vocem commoveat, sed quod nonnunquam in re dissimili parilis figurae formas in simili imponunt dispariles, ut calcei muliebres sint an viriles dicimus ad similitudinem figurae, cum tamen sciamus nonnunquam et mulierem habere calceos viriles et virum muliebris.

As to their asking from which part a word ought to be similar, from the voice or from the signification, we answer: from the voice; yet nonetheless we sometimes inquire whether the things signified are similar in kind, and we compare a virile noun with a virile, a female’s with a muliebral—not because that which they signify moves the voice, but because sometimes, in a dissimilar thing, they impose forms of like figure, and upon a similar thing, unlike ones; as when we say that shoes are muliebral or virile according to likeness of figure, although we know that sometimes a woman has virile shoes and a man muliebral ones.

Sic dici virum Perpennam ut Alfenam muliebri forma et contra parietem ut abietem esse forma similem, quom alterum vocabulum dicatur virile, alterum muliebre et utrumque natura neutrum sit. Itaque ea virilia dicimus non quae virum significant, sed quibus proponimus hic et hi, et sic muliebria in quibus dicere possumus haec aut hae.

Thus the man Perpenna is said in the virile form, as Alfenna in the feminine form; and conversely, the wall, as the fir-tree, is said to be similar in form, when the one vocable is called virile, the other feminine, and both are by nature neuter. Therefore we call those virile not which signify a man, but with which we set forth hic and hi, and likewise those feminine in which we can say haec or hae.

XXX.

30.

Quare nihil est, quod dicunt Theona et Diona non esse similis, si alter est Aethiops, alter albus, si analogia rerum dissimilitudines adsumat ad discernendum vocis verbi figuras.

Wherefore there is nothing in what they say, that Theon and Dion are not similar, if the one is an Aethiopian, the other white, if the analogy of things adopts dissimilarities for discerning the figures of the voice of the verb.

XXXI.

31.

Quod dicunt simile sit necne nomen nomini impudenter Aristarchum praecipere oportere spectare non solum ex recto, sed etiam ex eorum vocandi casu, esse enim deridiculum, si similes inter se parentes sint, de filiis iudicare: errant, quod non ab eorum obliquis casibus fit, ut recti simili facie ostendantur, sed propter eos facilius perspici similitudo potest eorum quam vim habeat, ut lucerna in tenebris allata non facit ut quae ibi sunt posita similia sint, sed ut videantur, quae sunt quoius modi sint.

As to what they say—that Aristarchus ought impudently to be instructed to look, not only from the nominative, but also from their vocative case, at whether a name is similar or not to a name—for it is derisory, if the parents are similar among themselves, to judge from the sons: they err, because it is not from their oblique cases that it comes about that the nominatives are shown with a similar face, but on account of them the similarity can more easily be perceived, what power it has; just as a lamp brought into the darkness does not make the things set there be similar, but makes them be seen—what they are, and of what sort they are.

Quid similius videtur quam in his est extrema littera crux Phryx? Quas, qui audit voces, auribus discernere potest nemo, cum easdem non esse similes ex declinatis verbis intellegamus, quod cum sit cruces et Phryges et de his extremis syllabis exemptum sit E, ex altero fit ut ex C et S crux, ex altero G et S Phryx. Quod item apparet, cum est demptum S: nam fit unum cruce, alterum Phryge.

What seems more similar than this—that in these the final letter is x: crux, Phryx? Which sounds no one who hears can distinguish by the ears, although we understand from the declined words that they are not alike; since, when it is cruces and Phryges and the E is taken away from these final syllables, from the one it comes about that from C and S there is crux, from the other that from G and S there is Phryx. The same is apparent when the S is removed: for the one becomes cruce, the other Phryge.

XXXII.

32.

Quod aiunt, cum in maiore parte orationis non sit similitudo, non esse analogian, dupliciter stulte dicunt, quod et in maiore parte est et si in minore parte sit, tamen sit, nisi etiam nos calceos negabunt habere, quod in maiore parte corporis calceos non habeamus.

What they say—that when in the greater part of the oration there is not similitude, there is no analogy—they say doubly foolishly, because both in the greater part it exists, and if it be in the lesser part, nevertheless it exists, unless they will also deny that we have shoes, because in the greater part of the body we do not have shoes.

XXXIII.

33.

Quod dicunt nos dissimilitudinem potius gratam acceptamque habere quam similitudinem: itaque in vestitu in supellectile delectari varietate, non paribus subuculis uxoris, respondeo, si varietas iucunditas, magis varium esse in quo alia sunt similia, alia non sunt: itaque sicut abacum argento ornari, ut alia paria sint, alia disparia, sic orationem.

That they say we rather have dissimilarity as pleasing and acceptable than similarity: and so in clothing and in household-furnishings we take delight in variety, not in matching under-tunics with a wife, I answer: if variety is jocundity, that is more various in which some things are similar, others are not: and so, just as an abacus is adorned with silver, so that some are pairs and others unequal, so too an oration.

Rogant, si similitudo sit sequenda, cur malimus habere lectos alios ex ebore, alios ex testudine, sic item genere aliquo alio. Ad quae dico non dissimilitudines solum nos, sed similitudines quoque sequi saepe. Itaque ex eadem supellectili licet videre: nam nemo facit triclinii lectos nisi paris et materia et altitudine et figura.

They ask, if similarity is to be followed, why we prefer to have couches, some of ivory, others of tortoise-shell, likewise also of some other kind. To which I say that we pursue not only dissimilarities, but similarities too, often. And so from the same furnishings one may see: for no one makes the couches of a triclinium unless matching both in material and in altitude and in figure.

Quis facit mappas triclinaris non similis inter se? Quis pulvinos? Quis denique cetera, quae unius generis sint plura?

Who makes triclinium table-napkins not similar to one another? Who cushions? Who, finally, the rest of the things which, being of one genus, are several?

Cum, inquiunt, utilitatis causa introducta sit oratio, sequendum non quae habebit similitudinem, sed quae utilitatem. Ego utilitatis causa orationem factam concedo, sed ut vestimenta: quare ut hic similitudines sequimur, ut virilis tunica sit virili similis, item toga togae, sic mulierum stola ut sit stolae proportione et pallium pallio simile, sic cum sint nomina utilitatis causa, tamen virilia inter se similia, item muliebria inter se sequi debemus.

“When,” they say, “since speech has been introduced for the sake of utility, we should follow not what will have similarity, but what has utility.” I grant that speech was made for the sake of utility, but as garments: wherefore, as here we follow likenesses—that a man’s tunic be similar to a man’s, likewise a toga to a toga; so the women’s stola should be in proportion to a stola, and the pallium like to a pallium—so, since names are for the sake of utility, nevertheless we ought to follow similarities, male things among themselves, likewise female things among themselves.

XXXIV.

34.

Quod aiunt ut persedit et perstitit sic periacuit et percubuit quoniam non sint, non esse analogian, et in hoc errant: quod duo posteriora ex prioribus declinata non sunt, cum analogia polliceatur ex duobus similibus similiter declinatis similia fore.

What they say—that just as persedit and perstitit, so periacuit and percubuit; since these are not, there is no analogy—and in this they err: because the two latter are not declined from the former, whereas analogy pledges that from two similar things, similarly declined, similar results will be.

XXXV.

35.

Qui dicunt quod sit ab Romulo Roma et non Romula neque ut ab ove ovilia sic a bove bovilia, non esse analogias, errant, quod nemo pollicetur e vocabulo vocabulum declinari recto casu singulari in rectum singularem, sed ex duobus vocabulis similibus casus similiter declinatos similes fieri.

Those who say that Rome is from Romulus and not Romula, and that, not as from a sheep “ovilia,” so from an ox “bovilia,” that there are no analogies, are in error, because no one promises that from a word another word is to be declined from the nominative case singular into the nominative singular, but that from two similar words the cases, declined in like manner, will be similar.

XXXVI.

36.

Dicunt, quod vocabula litterarum Latinarum non declinentur in casus, non esse analogias. Hi ea quae natura declinari non possunt, eorum declinatus requirunt, proinde et non eorum dicatur esse analogia quae ab similibus verbis similiter essent declinata. Quare non solum in vocabulis litterarum haec non requirenda analogia, sed ne in syllaba quidem ulla, quod dicimus hoc BA, huius BA, sic alia.

They say that, because the words of the Latin letters are not declined into cases, there are no analogies. These men require declensions of those things which by nature cannot be declined, and accordingly analogy is said not to belong even to those which would have been declined similarly from similar words. Wherefore this analogy is not to be required not only in the words of the letters, but not in any syllable at all, since we say “this BA,” “of this BA,” and so with others.

Quod si quis in hoc quoque velit dicere esse analogias rerum, tenere potest: ut enim dicunt ipsi alia nomina, quod quinque habeant figuras, habere quinque casus, alia quattuor, sic minus alia, dicere poterunt esse litteras ac syllabas in voce quae singulos habeant casus, in rebus pluris; quemadmodum inter se conferent ea quae quaternos habebunt vocabulis casus, item ea inter se quae ternos, sic quae singulos habebunt, ut conferant inter se dicentes, ut sit hoc A, huic A, esse hoc E, huic E.

But if someone should wish to say that there are analogies of things in this too, he can maintain it: for as they themselves say that some nouns, because they have five figures, have five cases, others four, thus others fewer, so they will be able to say that there are letters and syllables in the voice which have single cases, whereas in the things there are more; just as they will compare among themselves those which will have four by the names of cases, likewise those among themselves which will have three, so those which will have single, so that they compare among themselves saying, that this A, to this A; that this E, to this E.

XXXVII.

37.

Quod dicunt esse quaedam verba quae habeant declinatus, ut caput capitis, nihil nihili, quorum par reperiri quod non possit, non esse analogias, respondendum sine dubio, si quod est singulare verbum, id non habere analogias: minimum duo esse debent verba, in quibus sit similitudo. Quare in hoc tollunt esse analogias.

As to what they say, that there are certain words which have declensions, as caput capitis, nihil nihili, for which a matching pair cannot be found, that there are not analogies: it must without doubt be answered, that if a word is singular (unique), that does not have analogies; at a minimum there must be two words in which there is similarity. Wherefore in this they do away with analogies.

Sed nihilum vocabulum recto casu apparet in hoc:

But the vocable “nihilum” appears in the straight case in this:

Quae dedit ipsa capit neque dispendi facit hilum,
quod valet nec dispendii facit quicquam.

What she herself gave she takes, nor does she make a jot of loss,
what it is worth she makes no loss whatsoever.

Idem hoc obliquo apud Plautum:

The same in the oblique case in Plautus:

Video enim te nihili pendere prae Philolacho omnis homines,

For I see you count all men as nothing in comparison with Philolaches,

quod est ex ne et hili: quare dictus est nihili qui non hili erat. Casus tantum commutantur de quo dicitur, ut de homine: dicimus enim hic homo nihili et huius hominis nihili et hunc hominem nihili. Si in illo commutaremus, diceremus ut hoc linum et libum, sic nihilum, non hic nihili, et ut huic lino et libo, sic nihilo, non huic nihili.

which is from ne- and hīli: wherefore he is called nihili who was not hīli. Only the cases are changed of that about which it is spoken, as about a man: for we say “hic homo nihili” and “huius hominis nihili” and “hunc hominem nihili.” If we were to change it in that, we would say, as “hoc linum et libum,” so “nihilum,” not “hic nihili”; and as “huic lino et libo,” so “nihilo,” not “huic nihili.”

Potest dici patricus casus, ut ei praeponantur nomina plura, ut hic casus Terentii, hunc casum Terentii, hic miles legionis, huius militis legionis, hunc militem legionis.

It can be called the patrician case, since several nouns are set before it, as: this case of Terence, this case (accusative) of Terence; this soldier of the legion, of this soldier of the legion, this soldier (accusative) of the legion.

XXXVIII.

38.

Negant, cum omnis natura sit aut mas aut femina aut neutrum, non debuisse ex singulis vocibus ternas figuras vocabulorum fieri, ut albus alba album; nunc fieri in multis rebus binas, ut Metellus Metella, Aemilius Aemilia, nonnulla singula, ut tragoedus, comoedus; sic esse Marcum, Numerium, at Marcam, at Numeriam non esse; dici corvum, turdum, non, dici corvam, turdam; contra dici pantheram, merulam, non dici pantherum, merulum; nullius nostrum filium et filiam non apte discerni marem ac feminam, ut Terentium et Terentiam, contra deorum liberos et servorum non itidem, ut Iovis filium et filiam, Iovem et Iovam; item magnum numerum vocabulorum in hoc genere non servare analogias.

They deny that, since every nature is either male or female or neuter, it was not right that from single words triple figures of nouns be made, as albus alba album; now in many cases pairs are made, as Metellus Metella, Aemilius Aemilia, some single ones, as tragoedus, comoedus; thus there is Marcus, Numerius, but there is not Marca, nor Numeria; one says corvum, turdum, not corvam, turdam; conversely one says pantheram, merulam, not pantherum, merulum; for none of us are the son and daughter not aptly distinguished as male and female, as Terentium and Terentiam, but for the children of the gods and of slaves not in the same way, as the son and daughter of Jove, Iovem and Iovam; likewise a great number of nouns in this kind do not observe analogies.

Ad haec dicimus, omnis orationis quamvis res naturae subsit, tamen si ea in usum, non pervenerit, eo non pervenire verba: ideo equus dicitur et equa: in usu enim horum discrimina; corvus et corva non, quod sine usu id, quod dissimilis naturae. Itaque quaedam aliter olim ac nunc: nam et tum omnes mares et feminae dicebantur columbae, quod non erant in eo usu domestico quo nunc, et nunc contra, propter domesticos usus quod internovimus, appellatur mas columbus, femina columba.

To these points we say: although in all discourse some thing of nature underlies it, nevertheless, if that thing has not come into use, the words do not arrive there: for this reason “equus” is said and “equa”: for in these there are distinctions in use; “corvus” and “corva” are not said, because that, though of dissimilar nature, is without use. And so certain things were otherwise formerly than now: for even then both males and females were called “columbae,” because they were not in that domestic use in which they are now; and now contrariwise, on account of domestic uses, because we have come to know the difference, the male is called “columbus,” the female “columba.”

Natura cum tria genera transit et id est in usu discriminatum, tum denique apparet, ut est in doctus et docta et doctum: doctrina enim per tria haec transire potest et usus docuit discriminare doctam rem ab hominibus et in his marem ac feminam. In mare et femina et neutro neque natura maris transit neque feminae neque neutra, et ideo non dicitur feminus femina feminum, sic reliqua: itaque singularibus ac secretis vocabulis appellati sunt.

When nature transits the three kinds and that is discriminated in usage, then at last it appears, as it is in doctus and docta and doctum: for doctrine can transit through these three, and usage has taught to discriminate a learned thing from human beings, and among these the male and the female. In male and female and neuter neither does the nature of the male transit nor of the female nor of the neuter, and therefore one does not say feminus femina feminum, so with the rest: accordingly they have been appellated by singular and separate vocabularies.

Quare in quibus rebus non subest similis natura aut usus, in his vocabulis huiusce modi ratio quaeri non debet: ergo dicitur ut surdus vir, surda mulier, sic surdum theatrum, quod omnes tres res ad auditum sunt comparatae; contra nemo dicit cubiculum surdum, quod ad silentium, non ad auditum; at si fenestram non habet, dicitur caecum, ut caecus et caeca, quod omnia non habent quod lumen habere debent.

Wherefore, in those matters in which a similar nature or usage does not underlie, for words of this kind a rationale ought not to be sought: thus one says a deaf man, a deaf woman, and likewise a deaf theater, because all three things are referred to hearing; by contrast no one says a deaf bedroom, because it is for silence, not for hearing; but if it does not have a window, it is called blind, like a blind man and a blind woman, because in all these the thing they ought to have—light—is lacking.

Mas et femina habent inter se natura quandam societatem, nullam societatem neutra cum his, quod sunt diversa; inter se quoque de his perpauca sunt quae habeant quandam communitatem. Dei et servi nomina quod non item ut libera nostra transeunt, eadem est causa, quod ad usum attinet et institui opus fuit de liberis, de reliquis nihil attinuit, quod in servis gentilicia natura non subest in usu, in nostris nominibus qui sumus in Latio et liberi, necessaria. Itaque ibi apparet analogia ac dicitur Terentius vir, Terentia femina, Terentium genus.

The male and the female have between themselves by nature a certain society; the neuter has no society with these, because they are different; between themselves, moreover, of these there are very few which have any community. That the names of gods and of slaves do not pass in like manner as those of our freeborn has the same cause: so far as use is concerned, it was needful to be instituted about the freeborn; about the rest it did not matter, since in slaves a gentilician nature is not present in use, whereas in our names—who are in Latium and free—it is necessary. Therefore there the analogy appears, and one says Terentius a man, Terentia a woman, the Terentian lineage.

In praenominibus ideo non fit item, quod haec instituta ad usum singularia, quibus discernerentur nomina gentilicia, ut ab numero Secunda, Tertia, Quarta in mulieribus, in viris ut Quintus, Sextus, Decimus, sic ab aliis rebus. Cum essent duo Terentii aut plures, discernendi causa, ut aliquid singulare haberent, notabant, forsitan ab eo, qui mane natus diceretur, ut is Manius esset, qui luci, Lucius, qui post patris mortem, Postumus.

In praenomina it does not happen likewise, because these were instituted for singular use, by which the gentilician names might be distinguished, as from number “Secunda, Tertia, Quarta” among women, among men “Quintus, Sextus, Decimus,” so also from other things. When there were two Terentii or more, for the sake of distinguishing, so that they might have something singular, they would mark them—perhaps from this: that he who was said to be born in the morning would be Manius, he who at first light, Lucius, he who after the father’s death, Postumus.

E quibus aeque cum item accidisset feminis, proportione ita appellata declinarant praenomina mulierum antiqua, Mania, Lucia, Postuma: videmus enim Maniam matrem Larum dici, Luciam Volumniam Saliorum Carminibus appellari, Postumam a multis post patris mortem etiam nunc appellari.

From which, since equally the same likewise had happened for women, by proportion the ancient praenomina of women were thus declined and appellated: Mania, Lucia, Postuma; for we see “Mania, mother of the Lares” to be said, “Lucia Volumnia” to be named in the Songs of the Salii, and “Postuma” even now to be called by many after the father’s death.

Quare quocumque progressa est natura cum usu vocabuli, similiter proportione propagata est analogia, cum in quibus declinatus voluntarii maris et feminae et neutri, quae voluntaria, non debeant similiter declinari, sed in quibus naturales, sint declinatus hi qui esse reperiuntur. Quocirca in tribus generibus nominum inique tollunt analogias.

Wherefore, wherever nature has advanced together with the use of the vocable, likewise by proportion the analogy has been propagated; since, in those cases in which the inflections of the voluntary gender—masculine, feminine, and neuter—which are voluntary, ought not to be declined similarly, but, in those in which they are natural, let the inflections be those which are found to exist. Therefore, in the three genders of nouns they unjustly remove the analogies.

XXXIX.

39.

Qui autem eas reprehendunt, quod alia vocabula singularia sint solum, ut cicer, alia multitudinis solum, ut scalae, cum debuerint omnia esse duplicia, ut equus equi, analogiae fundamentum esse obliviscuntur naturam et usum. Singulare est quod natura unum significat, ut equus, aut quod coniuncta quodammodo ad unum usu, ut bigae: itaque ut dicimus una Musa, sic dicimus unae bigae.

But those who censure them, on the ground that some words are singular only, as cicer, others of multitude only, as scalae, whereas all ought to be double, as equus equi, forget that the foundation of analogy is nature and usage. The singular is what by nature signifies one, as equus, or what by usage is in some way conjoined to one, as bigae: and so, as we say one Muse, so we say one bigae.

Multitudinis vocabula sunt unum infinitum, ut Musae, alterum finitum, ut duae, tres, quattuor: dicimus enim ut hae Musae sic unae bigae et binae et trinae bigae, sic deinceps. Quare tam unae et uni et una quodammodo singularia sunt quam unus et una et unum; hoc modo mutat, quod altera in singularibus, altera in coniunctis rebus; et ut duo tria sunt multitudinis, sic bina trina.

Words of multitude are of two kinds: one infinite, as “Muses,” the other finite, as “two, three, four.” For we say, just as “these Muses,” so “one two-horse chariot” and “two two-horse chariots” and “three two-horse chariots,” and so on. Wherefore “unae” and “uni” and “una” are in a certain manner singular just as “unus” and “una” and “unum”; it changes in this way, that the one is in singulars, the other in conjoined things; and as “two” and “three” are of multitude, so “by twos” and “by threes.”

Est tertium quoque genus singulare ut in multitudine, uter, in quo multitudinis ut utrei; uter poeta singulari, utri poetae multitudinis est. Qua explicata natura apparet non debere omnia vocabula multitudinis habere par singulare: omnes enim numeri ab duobus susum versus multitudinis sunt neque eorum quisquam habere potest singulare compar. Iniuria igitur postulant, si qua sint singularia, oportere habere multitudinis.

There is also a third kind: singular as in plurality, uter, in which the plural is utrei; uter is of the singular, utri is of the plurality of poetae. With this nature explained, it appears that not all words of plurality ought to have a singular pair: for all numbers from two upward are of plurality, and none of them can have a singular counterpart. Unjustly, therefore, they demand that, if there are certain singulars, they ought to have a plurality.

XL.

40.

Item qui reprehendunt, quod non dicatur ut unguentum unguenta vinum vina sic acetum aceta garum gara, faciunt imperite: qui ibi desiderant multitudinis vocabulum, quae sub mensuram ac pondera potius quam sub numerum succedunt: nam in plumbo, argento, cum incrementum accessit, dicimus multum, sic multum plumbum, argentum; non plumba, argenta, cum quae ex hisce fiant, dicamus plumbea et argentea (aliud enim cum argenteum: nam id tum cum iam vas: argenteum enim, si pocillum aut quid item): quod pocilla argentea multa, non quod argentum multum.

Likewise those who find fault because it is not said, as unguentum–unguenta, vinum–vina, so also acetum–aceta, garum–gara, act unskillfully: they there desire a vocable of multitude for things which fall under measure and weights rather than under number; for in lead and silver, when an increment has come, we say “much,” thus “much lead, (much) silver,” not “leads, silvers,” whereas for the things that are made from these we say “leaden” and “argent” (for it is another matter with argenteum: for that is then when it is already a vessel; argenteum, for instance, if it is a little cup or something of the sort): because there are many silver little cups, not because there is much silver.

Ea, natura in quibus est mensura, non numerus, si genera in se habent plura et ea in usum venerunt, a genere multo, sic vina et unguenta, dicta: alii generis enim vinum quod Chio, aliud quod Lesbo, sic ex regionibus aliis. Aeque ipsa dicuntur nunc melius unguenta, cui nunc genera aliquot. Si item discrimina magna essent olei et aceti et sic ceterarum rerum eiusmodi in usu communi, dicerentur sic olea et aceta ut vina.

Those things, in whose nature there is measure, not number, if they have several genera within themselves and these have come into use, are named from the multiplicity of the genus, thus “wines” and “unguents”: for the wine that is from Chios is of one genus, that from Lesbos of another, and so from other regions. Likewise the very things are now more aptly called unguents, for which there are now several genera. If likewise there were great discrimina of oil and vinegar, and thus of other things of this sort in common use, they would be spoken of thus as “oils” and “vinegars,” as “wines.”

Quare in utraque re inique rescindere conantur analogias, et cum in dissimili usu similia vocabula quaerant et cum item ea quae metimur atque ea quae numeramus dici putent oportere.

Wherefore, in each of the two matters they iniquitously attempt to rescind the analogies, both when in dissimilar use they seek similar vocables, and likewise when they suppose that the things which we measure and the things which we enumerate ought to be said the same way.

XLI.

41.

Item reprehendunt analogias, quod dicantur multitudinis nomine publicae balneae, non balnea, contra quod privati dicant unum balneum, quom, plura balnea non dicant. Quibus responderi potest non esse reprehendendum, quod scalae et aquae caldae, pleraque cum causa, multitudinis vocabulis sint appellata neque eorum singularia in usum venerint; idemque item contra. Primum balneum (nomen est Graecum), cum introiit in urbem, publice ibi consedit, ubi bina essent coniuncta aedificia lavandi causa, unum ubi viri, alterum ubi mulieres lavarentur; ab eadem ratione domi suae quisque ubi lavatur balneum dixerunt et, quod non erant duo, balnea dicere non consuerunt, cum hoc antiqui non balneum, sed lavatrinam appellare consuessent.

Likewise they reprehend the analogies, because the public baths are called by a term of multitude “publicae balneae,” not “balnea,” whereas in contrast private persons say a single “balneum,” since they do not say several “balnea.” To whom it can be answered that this is not to be blamed, since “stairs” (scalae) and “hot waters” (aquae caldae), and many things with good cause, have been appellated by words of multitude, and their singulars have not come into use; and likewise conversely. The first balneum (the name is Greek), when it entered the city, took its public seat there where there were twin conjoined buildings for the sake of washing, one where men, the other where women were bathed; from the same reasoning, each man at his own house where he is bathed called it “balneum,” and, because there were not two, they were not wont to say “balnea,” whereas the ancients were accustomed to call this not “balneum,” but “lavatrina.”

Sic aquae caldae ab loco et aqua, quae ibi scateret, cum ut colerentur venissent in usum nostris, cum aliae ad alium morbum idoneae essent, eae cum plures essent, ut Puteolis et in Tuscis, quibus utebantur, multitudinis potius quam singulari vocabulo appellarunt. Sic scalas, quod ab scandendo dicuntur et singulos gradus scanderent, magis erat quaerendum, si appellassent singulari vocabulo scalam, cum origo nominatus ostenderet contra.

Thus “hot waters,” from the place and the water that gushed there, when they had come into use among our people to be frequented, since some were suitable for one ailment and others for another, and since they were several, as at Puteoli and in the Tuscan region, which they used, they called by a term of multitude rather than a singular vocable. Thus with “stairs,” because they are said from “to climb” and one would climb the individual steps, there would rather have been reason to question it if they had called it by a singular vocable, a stair, since the origin of the name would show the contrary.

XLII.

42.

Item reprehendunt de casibus, quod quidam nominatus habent rectos, quidam obliquos, quod dicunt utrosque in vocibus oportere. Quibus idem responderi potest, in quibus usus aut natura non subsit, ibi non esse analogiam ***

Likewise they find fault about cases, because certain names have straight cases, certain oblique, since they say that both ought to be in words. To these the same can be answered: where usage or nature is not present, there there is no analogy ***

Sed ne in his quidem vocabulis quae declinantur, si transeunt e recto casu in rectum casum: quae tamen fere non discedunt ab ratione sine iusta causa, ut hi qui gladiatores Faustini: nam quod plerique dicuntur, ut tris extremas syllabas habeant easdem, Cascelliani, Caeciliani, Aquiliani, animadvertant, unde oriuntur, nomina dissimilia Cascellius, Caecilius, Aquilius, Faustus: quod si esset Faustius, recte dicerent Faustianos; sic a Scipione quidam male dicunt Scipioninos: nam est Scipionarios. Sed, ut dixi, quod ab huiuscemodi cognominibus raro declinantur cognomina neque in usum etiam perducta, natant quaedam.

But not even in those vocables which are declined, if they pass from the direct case into the direct case: which, however, for the most part do not depart from reason without a just cause, as in those who are the gladiators of Faustus; for whereas most are so named as to have the same three last syllables—Cascelliani, Caeciliani, Aquiliani—let them observe whence they arise, from the dissimilar names Cascellius, Caecilius, Aquilius, Faustus: for if it were Faustius, they would rightly say Faustiani; thus from Scipio some wrongly say Scipionini: for it is Scipionarii. But, as I said, because from cognomina of this kind cognomina are rarely declined and have not even been brought into use, certain forms are afloat.

XLIII.

43.

Item dicunt, cum sit simile stultus luscus et dicatur stultus stultior stultissimus, non dici luscus luscior luscissimus, sic in hoc genere multa. Ad quae dico ideo fieri, quod natura nemo lusco magis sit luscus, cum stultior fieri videatur.

Likewise they say that, since stultus and luscus are similar, and one says “stultus, stultior, stultissimus,” one does not say “luscus, luscior, luscissimus,” and so with many things of this kind. To which I say this happens for the reason that by nature no one is more one‑eyed than a one‑eyed man, whereas someone seems able to become more stupid.

XLIV.

44.

Quod rogant, cur non dicamus mane manius manissime, item de vesperi: in tempore vere magis et minus esse non potest, ante et post potest. Itaque prius est hora prima quam secunda, non magis hora. Sed magis mane surgere tamen dicitur: qui primo mane surgit, magis mane surgit quam qui non primo: ut enim dies non potest esse magis quam dies, sic mane non magis quam mane; itaque ipsum hoc quod dicitur magis sibi non constat, quod magis mane significat primum mane, magis vespere novissimum vesper.

As to what they ask, why we do not say morning, more-morning, most-morning, likewise for evening: in the category of time there cannot be more and less; there can be before and after. And so the first hour is earlier than the second, not “more” an hour. Yet “more early (more morning)” is nevertheless said: he who rises at first morning rises more early than he who does not at first; for just as a day cannot be more than a day, so morning not more than morning; therefore this very thing which is said “more” is inconsistent with itself, because “more morning” signifies first morning, “more evening (vesper)” the latest evening.

XLV.

45.

Item ab huiuscemodi dissimilitudinibus reprehenditur analogia, quod cum sit anus cadus simile et sit ab anu anicula anicilla, a cado duo reliqua quod non sint propagata, sic non dicatur a piscina piscinula piscinilla. Ad haec respondeo huiuscemodi vocabulis analogias esse, ut dixi, ubi magnitudo animadvertenda sit in unoquoque gradu eaque sit in usu communi, ut est cista cistula cistella et canis catulus catellus, quod in pecoris usu non est. Itaque consuetudo frequentius res in binas dividi partis ut maius et minus, ut lectus et lectulus, arca et arcula, sic alia.

Likewise analogy is reproved from dissimilarities of this sort: because, although anus and cadus are similar, and from anus there are anicula and anicilla, from cadus the other two are not propagated; thus one should not say from piscina piscinula and piscinilla. To these I reply that for vocabula of this kind there are analogies, as I said, where magnitude is to be observed at each grade and it is in common use, as cista, cistula, cistella and canis, catulus, catellus—something which is not in the usage of livestock. Therefore custom more often divides things into two parts, as greater and lesser, as lectus and lectulus, arca and arcula, and so others.

XLVI.

46.

Quod dicunt casus alia non habere rectos, alia obliquos et ideo non esse analogias, falsum est. Negant habere rectos ut in hoc frugis frugi frugem, item colem colis cole, obliquos non habere ut in hoc Diespiter Diespitri Diespitrem, Maspiter Maspitri Maspitrem.

What they say, that some cases do not have straight (nominative) forms, others oblique, and therefore that there are no analogies, is false. They deny that there are straight forms, as in this frugis frugi frugem, likewise colem colis cole; that there are not oblique forms, as in this Diespiter Diespitri Diespitrem, Maspiter Maspitri Maspitrem.

Ad haec respondeo et priora habere nominandi et posteriora obliquos. Nam et frugi rectus est natura frux, at secundum consuetudinem dicimus ut haec avis, haec ovis, sic haec frugis; sic secundum naturam nominandi est casus cols, secundum consuetudinem colis, cum utrumque conveniat ad analogian, quod et id quod in consuetudine non est cuius modi debeat esse apparet, et quod est in consuetudine nunc in recto casu, eadem est analogia ac pleraque, quae ex multitudine cum transeunt in singulare, difficulter efferuntur ore. Sic cum transiretur ex eo quod dicebatur haec oves, una non est dicta ovs sine I, sed additum I ac factum ambiguum verbum nominandi an patrici esset casus.

To these things I reply that both the former have a nominative of naming and the latter have obliques. For frugi has by nature the straight (nominative) frux; but according to consuetude we speak as with “this avis,” “this ovis,” so “this frugis.” Thus according to the nature of naming the case is cols, according to consuetude colis, since each suits the analogy: both because that which is not in consuetude shows of what sort it ought to be, and because that which is in consuetude now in the straight case has the same analogy as very many words which, when they pass from the multitude into the singular, are with difficulty brought forth by the mouth. Thus, when one passed from what was being said “these oves,” the one was not said ovs without I, but an I was added, and the word was made ambiguous whether it was the case of naming or the patrius (i.e., the genitive) case.

Ut ovis, et avis.

As with ovis, and avis.

Sic in obliquis casibus cur negent esse Diespitri Diespitrem non video, nisi quod minus est tritum in consuetudine quam Diespiter; quod in nihil argumentum est: nam tam casus qui non tritus est quam qui est. Sed esto in casuum serie alia vocabula non habere nominandi, alia de obliquis aliquem: nihil enim ideo quo minus siet ratio percellere poterit hoc crimen.

Thus in the oblique cases I do not see why they deny the forms Diespitri, Diespitrem, unless because it is less worn in usage than Diespiter; which is an argument to nothing: for a case which is not worn is just as much a case as one that is. But be it so, that in the series of cases some vocables do not have the nominative, others have some among the obliques: for on that account nothing will be able to batter down the rationale, so that it be any the less, by this charge.

Nam ut signa quae non habent caput aut aliquam aliam partem, nihilo minus in reliquis membris eorum esse possunt analogiae, sic in vocabulis casuum possunt item fieri iacturae. Potest etiam refingi ac reponi quod aberit, ubi patietur natura et consuetudo: quod nonnunquam apud poetas invenimus factum, ut in hoc apud Naevium in Clastidio:

For just as statues which do not have a head or some other part nonetheless can have analogies in their remaining limbs, so too in the words of the cases losses can likewise occur. What is missing can also be refashioned and restored, where nature and usage will allow: which we sometimes find done among the poets, as in this instance in Naevius, in the Clastidium:

Vita insepulta laetus in patriam redux.

His life unburied, glad, redux to his fatherland.

XLVII.

47.

Item reprehendunt, quod dicatur haec strues, hic Hercules, hic homo: debuisset enim dici, si esset analogia, hic Hercul, haec strus, hic homen. Non haec ostendunt nomina non analogian esse, sed obliquos casus non habere caput ex sua analogia. Non, ut si in Alexandri statua imposueris caput Philippi, membra conveniant ad rationem, sic et Alexandri membrorum simulacro caput quod respondeat item sit?

Likewise they reprehend that it is said haec strues, hic Hercules, hic homo: for it ought to have been said, if there were analogy, hic Hercul, haec strus, hic homen. These do not show that nouns are without analogy, but that the oblique cases do not have a head according to their own analogy. Is it not, as if in the statue of Alexander you were to impose the head of Philip, would the limbs accord to proportion, so also in the simulacrum of Alexander’s members there should likewise be a head that corresponds?

Non, si quis tunicam in usu ita consuit, ut altera plagula sit angustis clavis, altera latis, utraque pars in suo genere caret analogia.

No: if someone, in practice, sewed a tunic in such a way that one panel had narrow clavi (stripes) and the other broad, each part, in its own kind, does not lack analogy.

XLVIII.

48.

Item negant esse analogias, quod alii dicunt cupressus, alii cupressi, item de ficis platanis et plerisque arboribus, de quibus alii extremum US, alii EI faciunt. Id est falsum: nam debent dici E et I, fici ut nummi, quod est ut nummis ficis, ut nummorum ficorum. Si essent plures ficus, essent ut manus; diceremus ut manibus, sic ficibus, et ut manuum, sic ficuum, neque has ficos diceremus, sed ficus, ut non manos appellamus, sed manus, nec consuetudo diceret singularis obliquos casus huius fici neque hac fico, ut non dicit huius mani, sed huius manus, nec hac mano, sed hac manu.

Likewise they deny that there are analogies, because some say cupressus, others cupressi, likewise concerning fig-trees, plane-trees, and very many trees, of which some make the final US, others EI. That is false: for they ought to be said with E and I, fici like nummi, that is: as nummis, ficis; as nummorum, ficorum. If there were more ficus, they would be like manus; we would say, as manibus, so ficibus, and as manuum, so ficuum; nor would we say these ficos, but ficus, just as we do not call them manos, but manus; nor would usage say in the singular the oblique cases huius fici nor hac fico, just as it does not say huius mani, but huius manus, nor hac mano, but hac manu.

XLIX.

49.

Etiam illud putant esse causae, cur non sit analogia, quod Lucilius scribit:

They also think this to be a cause why there is no analogy, because Lucilius writes:

Decusis,
Sive decusibus est.

Decusis,
Or it is with decusibus.

Qui errant, quod Lucilius non debuit dubitare, quod utrumque: nam in aere usque ab asse ad centussis numerus aes adsignificat, et eius numero finiti casus omnes ab dupondio sunt, quod dicitur a multis duobus modis hic dupondius et hoc dupondium, ut hoc gladium et hic gladius; ab tressibus virilia multitudinis hi tresses et "his tressibus confido," singulare "hoc tressis habeo" et "hoc tressis confido," sic deinceps ad centussis. Deinde numerus aes non significat.

Those are in error, since Lucilius ought not to have doubted, on both counts: for in bronze coinage, from the as up to the centussis, the number designates the bronze, and the cases limited by that number are all from dupondius, which is said by many in two ways, “hic dupondius” and “hoc dupondium,” as “hoc gladium” and “hic gladius”; from tressis, the masculine plurals of multitude, “hi tresses” and “his tressibus confido,” the singular “hoc tressis habeo” and “hoc tressis confido,” and so on up to centussis. Then the number does not designate the bronze.

Numeri qui aes non significant, usque a quattuor ad centum, triplicis habent formas, quod dicuntur hi quattuor, hae quattuor, haec quattuor; cum perventum est ad mille, quartum assumit singulare neutrum, quod dicitur hoc mille denarium, a quo multitudinis fit milia denarium.

The numbers which do not signify bronze (money), from four up to 100, have triple forms, because one says “hi quattuor, hae quattuor, haec quattuor” (masculine these four, feminine these four, neuter these four); when one comes to 1,000, it assumes a fourth form, a neuter singular, so that one says “hoc mille denarium” (this 1,000 of denarius-money), from which the plurality of multitude is “milia denarium” (thousands of denarii).

Quare quoniam ad analogias quod pertineat non opus est ut omnia similia dicantur, sed ut in suo quaeque genere similiter declinentur, stulte quaerunt, cur as et dupondius et tressis non dicantur proportione, cum as sit simplex, dupondius fictus, quod duo asses pendebat, tressis ex tribus aeris quod sit. Pro assibus nonnunquam aes dicebant antiqui, a quo dicimus assem tenentes "hoc aere aeneaque libra" et "mille aeris legasse."

Wherefore, since as regards analogies it is not necessary that all things be said alike, but that each in its own kind be declined in like manner, they foolishly ask why the as and the dupondius and the tressis are not said by proportion, since the as is simple, the dupondius fashioned, because it weighed two asses, the tressis from three of bronze, since it is. In place of asses the ancients sometimes said aes, whence we say, holding an as, “with this bronze and the brazen balance,” and “to have bequeathed a thousand of bronze.”

Quare quod ab tressis usque ad centussis numeri ex partibus eiusdem modi sunt compositi, eiusdem modi habent similitudinem: dupondius, quod dissimilis est, ut debuit, dissimilem habet rationem. Sic as, quoniam simplex est ac principium, et unum significat et multitudinis habet suum infinitum: dicimus enim asses, quos cum finimus, dicimus dupondius et tressis et sic porro.

Wherefore, since the numbers from the tressis up to the centussis are composed from parts of the same kind, they have a likeness of the same kind: the dupondius, which is dissimilar, as it ought, has a dissimilar rationale. Thus the as, since it is simple and a beginning, both signifies one and, as to multitude, has its own infinity: for we say asses, and when we make them definite, we say dupondius and tressis and so on.

Sic videtur mihi, quoniam finitum et infinitum habeat dissimilitudinem, non debere utrumque item dici, eo magis quod in ipsis vocabulis ubi additur certus numerus miliariis aliter atque in reliquis dicitur: nam sic loquontur, hoc mille denarium, non hoc mille denariorum, et haec duo milia denarium, non duo milia denariorum. Si esset denarii in recto casu atque infinitam multitudinem significaret, tunc in patrico denariorum dici oportebat; et non solum in denariis, victoriatis, drachmis, nummis, sed etiam in viris idem servari oportere, cum dicimus iudicium fuisse triumvirum, decemvirum, centumvirum, non triumvirorum, decemvirorum, centumvirorum.

Thus it seems to me, since the finite and the infinite have a dissimilarity, that both ought not to be said in the same way, all the more because in the very vocabularies, where a fixed number is added, in thousands it is said otherwise than in the rest: for thus they speak, “hoc mille denarium,” not “hoc mille denariorum,” and “haec duo milia denarium,” not “duo milia denariorum.” If denarii were in the straight (nominative) case and signified an infinite multitude, then in the oblique it ought to be said denariorum; and not only in denarii, victoriati, drachmae, coins, but even in men the same ought to be observed, when we say “iudicium fuisse triumvirum, decemvirum, centumvirum,” not “triumvirorum, decemvirorum, centumvirorum.”

Numeri antiqui habent analogias, quod omnibus est una regula, duo actus, tres gradus, sex decuriae, quae omnia similiter inter se respondent. Regula est numerus novenarius, quod, ab uno ad novem cum pervenimus, rursus redimus ad unum et VIIII; hinc et LXXXX et nongenta ab una sunt natura novenaria; sic ab octonaria, et deorsum versus ad singularia perveniunt.

The ancient numbers have analogies, in that for all there is one rule, two acts, three grades, six decuries, all of which correspond similarly among themselves. The rule is the novenary number, because, when we have reached from 1 to 9, we return again to 1 and 9; hence both 90 and nine hundred from one are of a novenary nature; thus from the octonary, and downward, they arrive at the singulars.

Actus primus est ab uno ad DCCCC, secundus a mille ad nongenta milia; quod idem valebat unum et mille, utrumque singulari nomine appellatur: nam ut dicitur hoc unum, haec duo, sic hoc mille, haec duo milia et sic deinceps multitudinis in duobus actibus reliqui omnes item numeri. Gradus singularis est in utroque actu ab uno ad novem, denarius gradus a decem ad LXXXX, centenarius a centum ad DCCCC. Ita tribus gradibus sex decuriae fiunt, tres miliariae, tres minores.

The first act is from one to 900, the second from 1,000 to 900,000; since the same obtained for one and for a thousand, each is called by a singular name: for as one says “this one,” “these two,” so “this thousand,” “these two thousands,” and thus onward of the multitude in the two acts all the remaining numbers likewise. The singular grade is in each act from one to nine, the denary grade from ten to 90, the centenary from 100 to 900. Thus by three grades six decuries come to be, three millenary, three lesser.

Antiqui his numeris fuerunt contenti.

The ancients were content with these numbers.

Ad hos tertium et quartum actum addentes ab deciens et ab deciens miliens minores imposuerunt vocabula, neque ratione, sed tamen non contra est eam de qua scribimus analogiam. Nam deciens dicatur hoc deciens ut mille hoc mille, ut sit utrumque sine casibus vocis, dicemus ut hoc mille, huius mille, sic hoc deciens, huius deciens, neque eo minus in altero, quod est mille, praeponemus hi mille, horum mille, sic hi deciens, horum deciens.

To these, adding a third and a fourth act, they imposed names for the numbers smaller from deciens and from deciens miliens, not by reason; yet nevertheless it is not contrary to the analogy about which we are writing. For let deciens be said “this deciens” just as mille is “this thousand,” so that both may be without cases of the word; we will say, as “this thousand, of this thousand,” so “this deciens, of this deciens,” nor any the less in the other, which is mille, we will prefix “these thousand, of these thousand,” so “these deciens, of these deciens.”

50.

Quoniam in eo est nomen commune, quam vocant homonymian, obliqui casus ab eodem capite, ubi erit homonymia, quo minus dissimiles fiant, analogia non prohibet. Itaque dicimus hic Argus, cum hominem dicimus, cum oppidum, Graece Graecaniceve hoc Argos, cum Latine hi Argi. Item faciemus, si eadem vox nomen et verbum significabit, ut et in casus et in tempora dispariliter declinetur, ut faciemus a Meto quod nomen est Metonis Metonem, quod verbum est metam metebam.

Since in it there is a common noun, which they call homonymy, analogy does not forbid the oblique cases to be from the same head, where there will be homonymy, in order that they may become less dissimilar. And so we say “this Argus” (hic Argus), when we mean a man; when we mean the town, in Greek or in Graecanic fashion “this Argos” (hoc Argos), whereas in Latin “these Argi” (hi Argi). Likewise we shall do, if the same voice signifies a noun and a verb, that it be declined differently both in cases and in tenses: thus we shall form from Meto, which is a name, “Metonis,” “Metonem”; which is a verb, “metam,” “metebam.”

LI.

51.

Reprehendunt, cum ab eadem voce plura sunt vocabula declinata, quas synonymias appellant, ut Alcmaeus et Alcmaeo, sic Geryon, Geryoneus, Geryones. In hoc genere quod casus perperam permutant quidam, non reprehendunt analogiam, sed qui eis utuntur imperite; quod quisque caput prenderit, sequi debet eius consequentis casus in declinando ac non facere, cum dixerit recto casu Alcmaeus, in obliquis Alcmaeoni et Alcmaeonem; quod si miscuerit et non secutus erit analogias, reprehendendum.

They reprehend, when from the same word several vocables are declined, which they call synonymies, as Alcmaeus and Alcmaeo, so Geryon, Geryoneus, Geryones. In this kind, when certain persons wrongly interchange the cases, they do not reprehend the analogy, but those who use them unskilfully; whichever head each has taken, he ought to follow its consequent cases in declining, and not do, when he has said in the nominative case Alcmaeus, in the oblique cases Alcmaeoni and Alcmaeonem; but if he has mixed them and has not followed the analogies, he is to be reprehended.

LII.

52.

Reprehendunt Aristarchum, quod haec nomina Melicertes et Philomedes similia neget esse, quod vocandi casus habet alter Melicerta, alter Philomedes, sic qui dicat lepus et lupus non esse simile, quod alterius vocandi casus sit lupe, alterius lepus, sic socer, macer, quod in transitu fiat ab altero trisyllabum soceri, ab altero bisyllabum macri.

They censure Aristarchus, because he denies that these names Melicertes and Philomedes are similar, on the ground that the vocative case has for the one Melicerta, for the other Philomedes; so too someone who would say that lepus and lupus are not similar, because the vocative of the one is lupe, of the other lepus; likewise socer and macer, because in transition it becomes from the one the trisyllabic soceri, from the other the bisyllabic macri.

De hoc etsi supra responsum est, cum dixi de lana, hic quoque amplius adiciam similia non solum a facie dici, sed etiam ab aliqua coniuncta vi et potestate, quae et oculis et auribus latere soleant: itaque saepe gemina facie mala negamus esse similia, si sapore sunt alio; sic equos eadem facie nonnullos negamus esse similis, si natione sunt ex procreante dissimiles.

Although this has been answered above, when I spoke about wool, here too I will add further that things are called similar not only by appearance, but also by some conjoined force and potency, which are wont to lie hidden both from eyes and from ears: and so we often deny apples of twin appearance to be similar, if they are of another savor; thus we deny some horses of the same appearance to be similar, if in breed they are unlike as to the sire.

Itaque in hominibus emendis, si natione alter est melior, emimus pluris. Atque in hisce omnibus similitudines non sumimus tantum a figura, sed etiam aliunde, ut in equis aetas, ut in asinis cuius modi faciant pullos, ut in pomis quo sint suco. Si igitur idem sequitur in similitudine verborum quis, reprehendundus non est.

Therefore, in buying human beings, if one is better by nation, we buy him for more. And in all these cases we take similarities not only from figure, but also from elsewhere: as in horses, age; as in asses, what sort of foals they produce; as in fruits, what juice they are of. If, therefore, someone follows the same in the similarity of words, he is not to be reprehended.

Quare similitudinum discernendarum causa nonnunquam ut pronomen assumitur, sic casum aliquem assumimus, ut in his nemus, lepus, hic lepus, hoc nemus: itaque discedunt ac dicuntur hi lepores, haec nemora. Sic aliud si quid assumptum erit extrinsecus, quo similitudo penitus perspici possit, non nimis erit remotum ab natura: neque enim magnetas lapides duo inter se similes sint necne, perspicere possis, nisi minutum extrinsecus prope apposueris ferrum, quod similes lapides similiter ducunt, dissimiliter dissimiles.

Wherefore, for the sake of discerning likenesses, just as a pronoun is sometimes assumed, so we assume some case, as in these: nemus, lepus, hic lepus, hoc nemus: and so they depart and are said: hi lepores, haec nemora. Likewise, if anything else has been assumed from without, whereby the similarity can be thoroughly perceived, it will not be too far removed from nature: for you cannot perceive whether two magnet-stones are similar to each other or not, unless you have set near from outside a minute piece of iron, which similar stones draw similarly, dissimilar ones dissimilarly.

Quod ad nominatuom analogiam pertinet, ita deliquatum arbitror, ut omnia quae dicuntur contra ad respondendum ab his fontibus sumi possit.

As far as the analogy of nouns is concerned, I judge it to be so clarified that everything which is said against it can, for responding, be taken from these fountains.

LIII.

53.

Quod ad verborum temporalium rationem attinet, cum partes sint quattuor, tempora, personae, genera, divisiones, ex omni parte quoniam reprehendunt, ad singula respondebo.

As far as the rationale of verb tenses is concerned, since the parts are four—tenses, persons, genera, divisions—and since they censure from every part, I will respond to each individually.

LIV.

54.

Primum quod aiunt analogias non servari in temporibus, cum dicant legi lego legam et sic similiter alia: nam quae sint ut legi rem perfectam significare, duo reliqua lego et legam inchoatam, iniuria reprehendunt: nam ex eodem genere et ex divisione idem verbum, quod sumptum est, per tempora traduci infecti potest, ut discebam disco discam, et eadem perfecti, ut didiceram didici didicero.

First, as to their saying that analogies are not observed in the tenses, when they say legi, lego, legam and likewise others similarly: for in that they hold that legi signifies a perfect thing, and the two remaining, lego and legam, an inchoate one, they censure unjustly: for from the same genus and by division the same verb, which has been taken, can be conveyed through the tenses of the imperfect (infectum), as discebam, disco, discam, and likewise of the perfect, as didiceram, didici, didicero.

LV.

55.

Ex quo licet scire verborum rationem constare, sed eos, qui trium temporum verba pronuntiare velint, inscienter id facere; item illos qui reprehendunt, quod dicamus amor amabor amatus sum: non enim debuisse in una serie unum verbum esse duplex, cum duo simplicia essent. Neque ex divisione si unius modi ponas verba, discrepant inter se: nam infecta omnia simplicia similia sunt, et perfecta duplicia inter se paria in omnibus verbis, ut haec amabar amor amabor, amatus eram amatus sum amatus ero.

From which it may be known that the rationale of verbs is consistent, but that those who wish to pronounce verbs of three tenses do this ignorantly; likewise those who find fault because we say I am loved I shall be loved I have been loved: for in one series one verb ought not to be double, when two are simple. Nor, on division, if you set verbs of one kind, do they disagree among themselves: for all the unfinished are simple and similar, and the finished, being double, are equal among themselves in all verbs, as these I was being loved I am loved I shall be loved, I had been loved I have been loved I shall have been loved.

Quare item male dicunt ferio feriam percussi, quod est ordo ferio feriam feriebam, percussi percussero percusseram. Sic deinceps in reliquis temporibus reprehendenti responderi potest.

Wherefore likewise they speak wrongly ferio feriam percussi, since the order is ferio feriam feriebam, percussi percussero percusseram. Thus, in turn, in the remaining tenses, an answer can be given to the critic.

LVI.

56.

Similiter errant qui dicunt ex utraque parte verba omnia commutare syllabas oportere aut nullum, in his pungo pungam pupugi, tundo tundam tutudi: dissimilia enim conferunt, verba infecti cum perfectis. Quod si infecta modo conferrent, omnia verbi principia incommutabilia viderentur, ut in his pungebam pungo pungam et contra ex utraque parte commutabilia, si perfecta ponerent, ut pupugeram pupugi pupugero.

Similarly they err who say that on both sides all verbs ought to commute syllables, or none, in cases like these pungo pungam pupugi, tundo tundam tutudi: for they are comparing dissimilar things, verbs of the Infectum with Perfects. But if they were comparing only the Infecta, all the beginnings of the verb would seem incommutable, as in these pungebam pungo pungam, and conversely commutable on both sides, if they were to set Perfects, as pupugeram pupugi pupugero.

LVII.

57.

Item male conferunt fui sum ero, quod fui est perfectum, cuius series sibi, ut debet, in omnibus partibus constat, quod est fueram fui fuero; de infectis sum quod nunc dicitur olim dicebatur esum et in omnibus personis constabat, quod dicebatur esum es est, eram eras erat, ero eris erit; sic huiusce modi cetera servare analogiam videbis.

Likewise they compare ill fui sum ero, because fui is perfect, whose series, as it ought, stands consistent with itself in all its parts, that is: fueram fui fuero; as for the infectives, sum, which now is said, once used to be said esum, and it was consistent in all persons, in that one said esum, es, est, eram, eras, erat, ero, eris, erit; thus you will see the rest of things of this kind preserve analogy.

LVIII.

58.

Etiam in hoc reprehendunt, quod quaedam verba neque personas habent ternas neque tempora terna: id imperite reprehendunt, ut si quis reprehendat naturam, quod non unius modi finxerit animalis omnis. Si enim natura non omnes formae verborum terna habent tempora, ternas personas, non habent totidem verborum divisiones. Quare cum imperamus, natura quod infecta verba solum habent, cum aut praesenti aut absenti imperamus, fiunt terna, ut lege legito legat: perfectum enim imperat nemo.

They also reprehend in this, that certain verbs have neither three persons nor three tenses: they reprehend that unskilfully, as if someone should reprehend nature because she has not fashioned all animals of one kind. For if by nature not all forms of verbs have three tenses and three persons, they do not have the same number of divisions of verbs. Wherefore, when we give commands, by nature verbs have only the unfinished tenses; since when we command one either present or absent, there come to be three, as: lege, legito, legat; for no one commands the perfect.

Contra quae sunt indicandi, ut lego legis legit, novena fiunt verba infecti, novena perfecti.

By contrast, those that are of the indicative, as lego legis legit, become nine forms of the imperfect, nine of the perfect.

LIX.

59.

Quocirca non si genus cum genere discrepat, sed in suo quique genere si quid deest, requirendum. Ad haec addita si erunt ea quae de nominatibus supra sunt dicta, facilius omnia solventur. Nam ut illic externis caput rectus casus, sic hic in forma est persona eius qui loquitur et tempus praesens, ut scribo lego.

Wherefore, it is not if a genus disagrees with a genus, but if in its own genus anything is lacking, that it is to be sought. To these, if there be added the things which have been said above about the nominatives, all things will be resolved more easily. For as there, in the externals, the head is the straight case (nominative), so here in the form there is the person of the one who speaks and the present time, as I write, I read.

Quare ut illic fit, si hic item acciderit, in formula ut aut caput non sit aut ex alieno genere sit, proportione eadem quae illic dicimus, cur nihilominus servetur analogia. Item, sicut illic caput suum habebit et in obliquis casibus transitio erit in aliam quam formulam, qua assumpta reliqua facilius possint videri verba, unde sint declinata (fit enim, ut rectus casus nonnunquam sit ambiguus), ut in hoc verbo volo, quod id duo significat, unum a voluntate, alterum a volando; itaque a volo intellegimus et volare et velle.

Wherefore, as it is done there, if it likewise should happen here, in the formula, that either the head be not, or be from an alien genus, then by the same proportion which we state there, let there be a reason why nevertheless analogy be observed. Likewise, just as there it will have its own head, and in the oblique cases there will be a transition into some other formula, by the assuming of which the remaining words can more easily be seen, whence they have been declined (for it happens that the straight case is sometimes ambiguous), as in this word volo, which signifies two things, one from voluntas (will), the other from volare (to fly); and so from volo we understand both to fly and to will.

LX.

60.

Quidam reprehendunt, quod pluit et luit dicamus in praeterito et praesenti tempore, cum analogiae sui cuiusque temporis verba debeant discriminare. Falluntur: nam est ac putant aliter, quod in praeteritis U dicimus longum pluit luit, in praesenti breve pluit luit: ideoque in lege venditionis fundi "ruta caesa" ita dicimus, ut U producamus.

Some reprove that we say pluit and luit in the past and in the present time, whereas the analogy ought to discriminate the verbs of each time. They are mistaken: for it is otherwise than they think, namely that in the past we say the U long—pluit, luit—, in the present short—pluit, luit; and therefore in the law of the sale of a farm, "ruta caesa," we so say it as to lengthen the U.

LXI.

61.

Item reprehendunt quidam, quod putant idem esse sacrifico et sacrificor, lavat et lavatur; quod sit an non, nihil commovet analogian, dum sacrifico qui dicat servet sacrificabo et sic per totam formam, ne dicat sacrificatur aut sacrificatus sum: haec enim inter se non conveniunt.

Likewise some reprehend, because they think sacrifico and sacrificor are the same, lavat and lavatur; whether this be so or not, it in no way moves the analogy, provided that he who says sacrifico preserves sacrificabo, and so through the whole form, lest he say sacrificatur or sacrificatus sum: for these do not agree among themselves.

Apud Plautum, cum dicit:

In Plautus, when he says:

Piscis ego credo qui usque dum vivunt lavant
Diu minus lavari quam haec lavat Phronesium,

I believe fish, who as long as they live lave,
are laved for less time than this woman laves Phronesium,

ad lavant lavari non convenit, ut I sit postremum, sed E; ad lavantur analogia lavari reddit: quod Plauti aut librarii mendum si est, non ideo analogia, sed qui scripsit est reprehendendus. Omnino et lavat et lavatur dicitur separatim recte in rebus certis, quod puerum nutrix lavat, puer a nutrice lavatur, nos in balneis et lavamus et lavamur.

with lavant, lavari does not agree; the last letter should be E, not I; with lavantur, analogy yields lavari: if it is a fault of Plautus or of the copyist, not therefore is analogy to be blamed, but he who wrote is to be reprehended. Altogether, both lavat and lavatur are said separately, rightly, in certain cases, in that the nurse washes the boy, the boy is washed by the nurse; we in the baths both wash and are washed.

Sed consuetudo alterum utrum cum satis haberet, in toto corpore potius utitur lavamur, in partibus lavamus, quod dicimus lavo manus, sic pedes et cetera. Quare e balneis non recte dicunt lavi, lavi manus recte. Sed quoniam in balneis lavor lautus sum, sequitur, ut contra, quoniam est soleo, oporteat dici solui, ut Cato et Ennius scribit, non ut dicit volgus, solitus sum, debere dici; neque propter haec, quod discrepant in sermone pauca, minus est analogia, ut supra dictum est.

But since usage, holding either alternative sufficient, rather employs lavamur for the whole body, lavamus for parts, in that we say “I wash my hands,” so the feet and the rest. Wherefore, from the baths they do not rightly say “I have washed” (lavi); “I have washed my hands” (lavi manus) is right. But since in the baths I am washed (lavor), I am bathed (lautus sum), it follows that, conversely, since there is soleo (“I am wont”), it ought to be said solui (“I was wont”), as Cato and Ennius write, not, as the vulgar crowd says, solitus sum (“I have been wont”), ought to be said; nor, because a few things are discrepant in speech, is analogy any the less, as was said above.

LXII.

62.

Item cur non sit analogia, afferunt, quod ab similibus similia non declinentur, ut ab dolo et colo: ab altero enim dicitur dolavi, ab altero colui; in quibus assumi solet aliquid, quo facilius reliqua dicantur, ut in Myrmecidis operibus minutis solet fieri: igitur in verbis temporalibus, quom similitudo saepe sit confusa, ut discerni nequeat, nisi transieris in aliam personam aut in tempus, quae proposita sunt non esse similia intellegitur, cum transitum est in secundam personam, quod alterum est dolas, alterum colis.

Likewise, why there is not analogy, they allege that from similars, similar forms are not declined, as from dolo and colo: from the one, for instance, it is said dolavi, from the other colui; in which it is wont that something be assumed, whereby the remaining forms may be said more easily, as is done in the minute works of Myrmecides: therefore in temporal verbs, since the similitude is often confounded, so that it cannot be discerned unless you have crossed over into another person or into another tense, it is understood that the things proposed are not similar, when there has been a transition into the second person, because the one is dolas, the other colis.

Itaque in reliqua forma verborum suam utrumque sequitur formam. Utrum in secunda persona forma verborum temporalium habeat in extrema syllaba AS an ES an IS aut IS, ad discernendas similitudines interest: quocirca ibi potius index analogiae quam in prima, quod ibi abstrusa est dissimilitudo, ut apparet in his meo, neo, ruo: ab his enim dissimilia fiunt transitu, quod sic dicuntur meo meas, neo nes, ruo ruis, quorum unumquodque suam conservat similitudinis formam.

Therefore, in the remaining shape of verbs each follows its own form. Whether in the second person the form of temporal verbs has in the final syllable AS or ES or IS or IS, matters for discerning similitudes: wherefore there rather is an index of analogy there than in the first, because there the dissimilitude is abstruse, as appears in these meo, neo, ruo: for from these unlike forms arise in the transition, since they are thus said meo meas, neo nes, ruo ruis, each of which preserves its own form of similitude.

LXIII.

63.

Analogiam item de his quae appellantur participia reprehendunt multi; iniuria: nam non debent dici terna ab singulis verbis amaturus amans amatus, quod est ab amo amans et amaturus, ab amor amatus. Illud analogia quod praestare debet, in suo quicque genere habet, casus, ut amatus amato et amati amatis; et sic in muliebribus amata et amatae; item amaturus eiusdem modi habet declinationes, amans paulo aliter; quod hoc genus omnia sunt in suo genere similia proportione, sic virilia et muliebria sunt eadem.

Many also reprehend analogy with respect to those things which are called participles; unjustly: for three should not be said to come from each single verb—amaturus, amans, amatus—since from amo come amans and amaturus, from amor amatus. That which analogy ought to furnish, it has in each one’s own kind, the cases, as amatus amato and amati amatis; and so in the feminine, amata and amatae; likewise amaturus has declinations of the same mode, amans a little otherwise; because in this class all things are similar in proportion in their own kind, thus the masculine and the feminine are the same.

LXIV.

64.

De eo quod in priore libro extremum est, ideo non esse analogiam, quod qui de ea scripserint aut inter se non conveniant aut in quibus conveniant ea cum consuetudinis discrepent verbis, utrumque est leve: sic enim omnis repudiandum erit artis, quod et in medicina et in musica et in aliis multis discrepant scriptores; item in quibus conveniunt in scriptis, si ea tamen repudiat natura: quod ita ut dicitur non sit ars, sed artifex reprehendendus, qui dici debet in scribendo non vidisse verum, non ideo non posse scribi verum.

Concerning that which stands at the end of the prior book, namely that therefore there is no analogy because those who have written about it either do not agree among themselves, or, in those points in which they do agree, those things disagree with the words of usage—both are trivial: for thus every art would have to be repudiated, since in medicine and in music and in many other fields the writers disagree; likewise, in those points in which they agree in their writings, if nevertheless nature repudiates them. As it is thus said, it is not the art that fails, but the artificer who is to be reprehended, who ought to be said, in writing, not to have seen the true—not therefore that the true cannot be written.

Qui dicit hoc monti et hoc fonti, cum alii dicant hoc monte et hoc fonte, sic alia quae duobus modis dicuntur, cum alterum sit verum, alterum falsum, non uter peccat tollit analogias, sed uter recte dicit confirmat; et quemadmodum is qui peccat in his verbis, ubi duobus modis dicuntur, non tollit rationem cum sequitur falsum, sic etiam in his quae non duobus dicuntur, si quis aliter putat dici oportere atque oportet, non scientiam tollit orationis, sed suam inscientiam denudat.

He who says “hoc monti” and “hoc fonti,” while others say “hoc monte” and “hoc fonte,” and likewise in other things that are said in two ways, when the one is true and the other false, it is not the one who errs that removes the analogies, but the one who speaks rightly that confirms them; and just as he who errs in these words, where they are said in two ways, does not remove the rationale when he follows the false, so also in those which are not said in two ways, if someone thinks they ought to be said otherwise than they ought, he does not remove the science of speech, but denudes his own ignorance.

LXV.

65.

Quibus rebus solvi arbitraremur posse quae dicta sunt priori libro contra analogian, ut potui brevi percucurri. Ex quibus si id confecissent quod volunt, ut in lingua Latina esset anomalia, tamen nihil egissent ideo, quod in omnibus partibus mundi utraque natura inest, quod alia inter se similia, alia dissimilia sunt, sicut in animalibus dissimilia sunt, ut equus bos ovis homo, item alia, et in uno quoque horum genere inter se similia innumerabilia. Item in piscibus dissimilis muraena lupo, is soleae, haec muraenae et mustelae, sic aliis, ut maior ille numerus sit similitudinum earum quae sunt separatim in muraenis, separatim in asellis, sic in generibus aliis.

By means of which matters I have, as I could, briefly run through how the things that were said in the prior book against analogy might be resolved. From which, even if they had accomplished what they want, namely that in the Latin language there is anomaly, nevertheless they would have achieved nothing for this reason: that in all parts of the world both natures are present—that some things are similar among themselves, others dissimilar—just as among animals there are dissimilars, as horse, ox, sheep, man, and others likewise; and in each of these kinds there are, among themselves, innumerables that are similar. Likewise among fishes the muraena is dissimilar to the lupus, that one to the sole, this one to the muraena and the mustela; so with others, so that the greater number is of the similarities of those things which are, separately, in muraenae, separately in aselli, so in other genera.

Quare cum in inclinationibus verborum numerus sit magnus a dissimilibus verbis ortus, quod etiam vel maior est in quibus similitudines reperiuntur, confitendum est esse analogias. Itemque cum ea non multo minus quam in omnibus verbis patiatur uti consuetudo communis, fatendum illud quoquo modo analogian sequi nos debere universos, singulos autem praeterquam in quibus verbis offensura sit consuetudo communis, quod ut dixi aliud debet praestare populus, aliud e populo singuli homines.

Wherefore, since in the inflections of words the number is great, arisen from dissimilar words—indeed perhaps even greater in those in which likenesses are found—it must be confessed that there are analogies. Likewise, since the common custom permits us to use it in well-nigh all words, we must admit that, in any case, we all collectively ought to follow analogy, but individuals should except those words in which the common custom would take offense; for, as I said, the people ought to furnish one thing, and individual men from the people another.

Neque id mirum est, cum singuli quoque non sint eodem iure: nam liberius potest poeta quam orator sequi analogias. Quare cum hic liber id quod pollicitus est demonstraturum absolverit, faciam finem; proxumo deinceps de declinatorum verborum forma scribam.

Nor is that surprising, since individuals too are not under the same law: for a poet can follow analogies more freely than an orator. Therefore, when this book shall have finished demonstrating what it has promised, I will make an end; next, thereafter I will write about the form of words that are declined.