Marcus Terentius Varro’s On the Latin Language is a first-century-B.C. treatise on the origins and meanings of Latin words, written by a Roman scholar who had served Pompey, been pardoned by Caesar, and survived proscription to die in 27 B.C. in his eighty-ninth year. Of its original twenty-five books, only Books V through X survive, and this Loeb volume presents the first three of those—Books V, VI, and VII—in Latin with facing English translation by Roland G. Kent. Varro composed the work in 47–45 B.C. and published it before Cicero’s death; Book I, now lost, was an introduction dedicated to Cicero, and the surviving Books V–VII open with a fresh dedication to him.
Varro divides the study of words into four questions: why and whence names were imposed on things, how they were imposed, why they change, and how they are used. Books V–VII take up the first of these—the origin of names—and Varro announces at the start of Book V that he will treat places and the things in them, then time, then the rare words of the poets.
The Making of a Survivor
Marcus Terentius Varro was born in 116 B.C., probably at Reate in the Sabine country, into an equestrian family. He studied under the grammarian Lucius Aelius Stilo and later took up philosophy at Athens with Antiochus of Ascalon. His public life reads like a map of the late Republic’s wars: he served against the rebel Sertorius in Spain in 76, held an officer’s post under Pompey against the Cilician pirates in 67, and presumably joined Pompey’s campaign against Mithradates. When civil war came he stood with Pompey in Spain, Epirus, and Thessaly. Caesar pardoned him, and he settled quietly at Rome, where he was appointed librarian of Caesar’s planned collection of Greek and Latin books. After the assassination he was proscribed by Antony, and his villa at Casinum, library and all, was destroyed; friends’ devotion got him away, Octavian’s protection secured him, and he died in 27 B.C. in his eighty-ninth year. Seventy-four works and some six hundred and twenty books came from him, covering nearly every field of knowledge. Only the De Re Rustica survives whole; the De Lingua Latina, twenty-five books in all, comes down as a torso, with just Books V–X extant and serious gaps between them.
The Shape of the Treatise
The grammatical works are catalogued in turn: the De Lingua Latina itself, the De Antiquitate Litterarum in two books addressed to Lucius Accius, the De Origine Linguae Latinae in three books to Pompey, a work On Characters, five books of Plautine Questions, three On the Similarity of Words, at least four On the Usefulness of Language, five or more On Latin Speech to Marcellus, and the nine books of Disciplinae. The great treatise was composed in 47–45 B.C. and published before Cicero’s death in 43. Book I was an introduction dedicating the work to Cicero; the rest falls into four sections of six books each. Books II–IV handled the imposition of words, probably drawing on an earlier smaller work on etymology dedicated to a quaestor of Varro’s named Septumius: the second book gave the arguments against etymology, the third those in its favour, the fourth its nature. Books V–VII open with a fresh dedication to Cicero—Book V on the names of places and the objects found there, Book VI on words for time, verbs above all, Book VII on the rare and difficult words of the poets. Books VIII–XIII turned to derivation and to the quarrel between Anomaly, which rests on usage, and Analogy, which rests on rational principle: the eighth book against Analogy, the ninth for it, the tenth Varro’s own solution, granting that analogy exists; the eleventh through thirteenth applied it. Books XIV–XIX treated syntax, and the last six books probably went on to stylistic and rhetorical embellishment. Of all this only V–X survive, and fragments of the rest turn up in Aulus Gellius.
Varro Among the Grammarians
On linguistic theory the introduction is measured: Varro owed much to his teacher Stilo, drew heavily on Greek predecessors, and followed neither Anomalists nor Analogists to extremes. His instinct was to derive Latin words from Latin sources rather than refer nearly everything to Greek origins—a preference the editor notes and then hands over to the specialists, Barwick, Kowalski, Dam, Dahlmann, Kriegshammer, Muller, Wölfflin.
The Manuscripts and the Editor’s Task
The text rests chiefly on the Codex Laurentianus li. 10, a parchment in Langobardic characters from the eleventh century, held at Florence and containing folios 2 to 86. Petrus Victorius and Iacobus Diacetius examined it in 1521 and collated it against a copy of the first printed edition, recording the differences; their copy survives at Munich and has manuscript value for the stretch from V. 119 to VI. 61, where a quire has since been lost from the Laurentian—readings known as Fv, with the first edition’s readings where no variation is recorded marked (Fv). A single folio of Codex Cassinensis 361, also eleventh century, preserves a fragment covering V. 41. The editor insists that a text facing a translation must be intelligible: corrupt passages emended, gaps filled. Much can be restored because Festus, Nonius Marcellus, and others quoted Varro nearly verbatim, and their testimony clears up obscure places. Departures from the manuscripts, violent as they look, are required as a basis for translation, though the text stays conservative; italics and pointed brackets, with the apparatus, show what actually stands in the codex. The apparatus names the emending scholar first, then the manuscript reading, working back through successive emendations to the source. Translation posed its own problems: Varro constantly explains one Latin word by another, so Latin must remain in the English, glossed with equivalents; technical matter demands close adherence; metrical citations are rendered in the same metre. Only two modern translations existed before—Huot’s French paraphrase, which often drops whole sentences, and Canal’s Italian; there was none in German or English. The notes cover technical linguistics, historical and geographical allusions, public and private life, odd word-forms, false etymologies flagged as incorrect, the sources of quotations, and the names of metres. Citations follow fixed schemes: Festus by page and line in Mueller’s Leipzig edition of 1839, the Grammatici Latini by volume, page, and line in Keil, Nonius by page and line in Mercier. Symbols are explained: additions in angle brackets, changed letters in italics, with certain routine normalizations—a for e, o for ae, b for u—left unrecorded.
The Heavens, Latium, and the Measure of Fields
Book V opens with the heavens. Varro notes that what we call the Dipper the Americans usually name otherwise, and the editor observes that the points of the compass are here, as often in the ancients, somewhat distorted. Pythagoras of Tarentum, Varro writes, made a bronze group of the two—the editor supplying the missing noun. Many nations inhabit Europe’s localities, named either by names transferred from men, like the Sabini and Lucani, or derived from men, like Apulia and Latium, or by both at once, like Etruria and Tusci. Where Latinus held his kingdom the whole field-land is called Latian, but piecemeal it takes its name from towns—Praenestine from Praeneste, Arician from Aricia.
Plants and Trees Without Breath
Varro moves to things said to live without breath — bushes and vines. He ties virgultum to viridis, and viridis to a “power” (vis) of moisture: let it dry out and the bush dies. Vitis comes from vinum, the source of wine. Malum he borrows from Aeolic Greek. Iuglans joins Iuppiter (Optimus Maximus) to glans, since the inner nut is the best and biggest of its kind; the same nut is nux because its juice blackens the skin as nox blackens the air. The editor’s footnote is blunt: every etymology here is wrong except malum and iuglans.
Garden Plants, Greek and Native
Garden plants carry foreign names. Ocimum, menta, ruta (now πήγανον), caulis, lapathium, and radix — the old Greek word for what is now ῥάφανος — are listed alongside serpyllum and rosa, the latter changed by a single letter. Latin taken from Greek includes coriandrum, malache, cuminum; lilium from λείριον, malva from μαλάχη, sisymbrium from σισύμβριον. The editor notes that all are Greek loans except radix, rosa, and malva. Native words follow: lactuca from lac, since it contains milk; brassica as praesica, from stalks cut off; asparagi from the rough aspera bushes, unless from Greek ἀσπάραγος; cucumeres from their curvature (curvimeres); fructus from ferre, the things the farm bears for enjoyment, whence fruges and frumentum — the latter because grits (mola, salt mixed with ground spelt) are added to boiled vitals; uvae from uvor, moisture.
Food, Bread, and Porridge
Man-made things come next: food, clothing, tools. Puls, porridge, is the oldest — named from Greek, or, per Apollodorus, from the sound it makes dropped into boiling water. Panis was first shaped like a weaver’s panus; later shapes gave panificium (from panis and facere), hence panarium, a bread-closet, like granarium, a grain-store — unless from Greek κρανός. Hordeum from horridus, bristling; triticum because tritum, threshed, from the ears; far from facere, made into flour at the mill; milium from Greek μελίνη; libum because a portion was offered (libabatur) before eating; testuacium baked in a heated testu pot, as matrons still do at the Matralia; circuli from the circuitus of the batter — flour, cheese, water. Careless makers called them lixulae or similixulae, a Sabine word; globi from a globus of dough cooked in oil; crustulum from the crusta of porridge, burnt like a corium. Others, thrion and placenta, are Greek.
From Porridge to Meat
Pulmentum is what was eaten with puls, as Plautus has it; hence pulmentarium. Caseus from milk coactum (coaxeus). After men stopped eating fireless natural foods, they boiled down fruits in an olla — hence holera; rapa from being dug (eruitur) from the earth; olea from ἐλαία; orchitis, the large olive, so called by the Athenians. On meat: suilla from sus; men began with roasted, then boiled, then cooked in juice. Assum because it assudescit, sweats, from the fire; elixum from the liquor of water; ex iure because more iucundum with seasoning. Succidia from cutting up swine, the first animal slaughtered and salted; tegus from tegere; perna from pes; sueris from the animal’s name; offula from offa; insicia from insecta caro, echoed in the Song of the Salii (prosicium, prosectum); murtatum from murta added to stuffings. Lucanica — soldiers learned it from the Lucanians — like Faliscan venter from Falerii; fundolus, open at one end, hence the Greek “blind intestine”; farcimina from stuffing; farticulum; hila from hilum, with Ennius quoted: “neque dispendi facit hilum”; apexabo from its apex-like projection; longavo because longer.
Offerings, Wool, and Clothing
Augmentum is the piece put on the victim’s liver for increasing; magmentum from magis, piety, hence magmentaria fana; mattea from Greek ματτύη; other Greek items include ovum and bulbum. Lana is Greek, per Polybius and Callimachus; purpura from the color of the sea purple-fish, a Punic word first brought by the Phoenicians; stamen from stando; subtemen, which goes under the warp; trama, where the cold passes through; densum from the comb’s teeth; filum because it is the smallest hilum. Pannus is Greek, E changed to A; panuvellium from panus and volvere; tunica from tuendo, as tuendica; toga from tegere; cinctus and cingillum from cingere, men’s and women’s respectively. Arma from arcere; parma, even from the center; conum, narrowed to a point; hasta, standing; iaculum, for throwing.
Implements of the Farm
Harrows (irpices)
Saying, Teaching, and the Roots of Speech
Varro works through dico, tracing it to the Greek deiknymi, “I show.” From this root he spins out a whole family: dicare, to show or dedicate (Ennius is cited on the circus’s six turning-posts); iudicare, to judge, since ius, right, is dicitur, spoken; index, the judge who speaks the law; dedicare, to dedicate, because a temple is consecrated when the magistrate speaks the fixed formulas after the pontiff. The derivatives multiply—iudicium, indicit for declaring war, indixit for inviting to a funeral, prodixit for postponing a day, addixit for awarding a decision. A dictum is a bon mot in farce, a man is dictiosus for wit, dicta are the leaders’ orders in camp, dictata school exercises, and the dictator is so called because he must be dictus, named by the consul. Doceo, to teach, comes from dico joined with trado—handing over what the learner did not know—or because teaching is saying, or because learners are inducuntur, led on; dux and ductor follow from ducere, and disciplina and discere from the same root by a change of letters. Disputatio and computatio both come from putare, to make clean: the putator trims trees, accounts are putari when the sum is net, and speech is disputare when words are neatly arranged. Disserit is borrowed from gardening—as a gardener lays out his plots, so the clear speaker is disertus—and sermo comes from series, a joining, whence serta garlands and sartum in clothing, since conversation requires another person joined with one. Hence conserere manum, hand-to-hand fighting, and the augur’s formula about the sacred bough and the consortes.
Lots, Reading, and the Sounds of Men
Sors, a lot, yields consortes and sortes, joining times with men and things, along with sortilegi and sors as principal at interest. Legere, to pick or read, covers letters picked with the eyes, legati chosen to be sent, leguli who pick olives and grapes, legumina, and leges, chosen and brought before the people, with legitima, collegae, sublecti, allecti, collecta, legio, diligens, and dilectus trailing after. Words for noise follow the likeness of sound—murmurari from the poets’ murmuring sea-shore, and fremere, gemere, clamare, crepare in the same manner. Quiritare is shouting for the protection of the Quirites, distinct from the rustic iubilare; the Quirites descend from the Curenses who came with King Tatius into the state. Triumphare comes from the soldiers shouting as the general returns through the City to the Capitol, though Varro allows it may derive from the Greek thriambos, a surname of Liber.
Pledges, Hope, and the Bondsman
Spondere derives from sponte, meaning will; Lucilius and Terence are quoted, and the derivatives follow—despondet, respondet, desponsor, sponsa, sponsor, consponsus, with Naevius’s consponsi and the betrothal terms sponsio and sponsalis. To despondisse filiam is to give a daughter who leaves her father’s sponte and is bound by the sponsus even unwillingly; despondisse animum gives us the despondent. Respondere answers the asker’s spons, and joking promises are not actionable. Spes, hope, comes from sponte, hoping what one wishes, with Plautus quoted on “my hope.” The sponsor, praes, and vas differ: the praes guarantees to the state when questioned by a magistrate, the vas promises bond for another, and custom gave a bondsman when a party seemed unreliable.
The Augur’s Invitation and the Turn to Greek
At the close of Book VI, Varro explains that the augur now stands with the consul when the citizen-army is summoned, reciting the formulas in advance; the consul orders the augur—not the assistant or herald—to call the inlicium, the invitation. He believes this began when the assistant was absent, that it made no difference to whom the order was given, and that certain things are done for form’s sake without always being said or done the same way. He found inlicium in the Commentaries of Marcus Junius; in Plautus’s Persa, inlex is one who does not obey the lex, and illex is that which entices. Having spoken at length on few matters, he promises brevity on many—especially Latin words thought to be Greek: scalpere from skaleuein, sternere from stronnynai, lingere from lichmasthai, i from ithi, ite from ite, gignitur from gignetai, ferte from pherete, and a long list besides, down to runcinare from the Greek rhykane. He stops, noting he undertook three books on these topics—two on prose, one on poetry—and has sent the two prose books, on places and on time; next he will begin on poetic word-origins. Book VI ends, and Book VII opens with a leaf missing from the model copy.
Poetic Words and the Buried Meanings
Poetic words are hard to expound, their meanings buried by time or obscured by lost letters. Varro defends the etymologist who adds or removes letters, comparing him to the viewer who sets black hairs behind ivory to see Myrmecides’ carving. He cites Aelius on the Salian Hymns, Epimenides waking after fifty years, and Teucer unrecognized after fifteen, and argues one should favor the diligent investigator, since not every origin can be given—as medicine’s efficacy or a tree’s roots may be unknown.
Etymology as a Theory of Culture
What holds the three surviving books together is not linguistic accuracy—Varro’s derivations are wrong far more often than right, and the translator’s notes say so bluntly—but a conviction that a word, properly opened, discloses the thing it names and the life that named it. When Varro derives iugerum from two joined actus, or sestertius from semis tertius, or Septimontium from the seven hills, he is not merely guessing at sounds; he is reconstructing a Roman world of fields measured by oxen, coins counted in asses, and festivals tied to hills and crossroads. Etymology becomes a form of antiquarianism. The gods receive the same treatment: Jupiter, Juno, Sol, Luna, Neptune, Mars, and Quirinus are each resolved into a natural force or a social function, so that the Roman pantheon emerges as a coded account of sky, earth, sea, and war. Varro’s method is openly comparative—Greek is invoked constantly, sometimes to derive a Latin word, sometimes to be rejected in favour of a native root—and he is candid that many origins cannot be recovered at all, comparing the search to asking why a medicine cures or where a tree’s roots begin.
The Poet’s Language and the Limits of Recovery
Book VII, the last in this volume, turns to words that the poets use and that ordinary speech has buried. Here Varro’s difficulties multiply, and he says so: meanings are obscured by time and by lost letters, and he defends the etymologist who adds or removes a letter the way one might set black hairs behind ivory to see a craftsman’s carving. The book works through the vocabulary of ritual and augury—templum in its three senses, the ancilia that fell from heaven in Numa’s reign, the twenty-seven rush figures called Argei thrown yearly into the Tiber—alongside the language of early Latin poetry, from Ennius and Naevius to Pacuvius, Accius, and Lucilius. Varro quotes them as evidence, and in doing so preserves lines that would otherwise be lost. The volume’s text breaks off where a leaf is missing from the model manuscript, a reminder that what survives of Varro is itself a torso.
A Scholar Between Greek and Latin
Varro’s place in the history of Latin letters rests on more than this one work. He wrote seventy-four works totaling some six hundred and twenty books, ranging across agriculture, law, religion, and grammar; only the De Re Rustica survives complete. His teacher Aelius Stilo shaped his habits of interpretation, and he drew heavily on Greek predecessors while resisting the extreme positions of both the Anomalists, who held language to be governed by usage and irregularity, and the Analogists, who held it to be governed by rule and ratio. In the books that follow these, Varro would take up that quarrel directly and propose his own solution. The De Lingua Latina reached later ages in fragments—quoted by Festus, Nonius Marcellus, Aulus Gellius, and others—and its manuscript tradition hangs largely on a single eleventh-century codex from Florence, collated by Renaissance scholars and edited repeatedly since. That such a battered text still transmits Varro’s patient, wrongheaded, endlessly curious account of how Romans came to call things what they called them is the small miracle this edition exists to serve.
