Synonyms in Latin and the Thesaurus Linguae Latinae

Guest Blog by Johann Ramminger

„Language possesses no pure synonyms. […] If, for extralinguistic reasons such as cultural borrowing or foreign conquest, two exact synonyms exist for a time in the spoken chain, either one of them will become less and less selected and eventually discarded, or a difference of meaning, connotation, nuance or register will arise to distinguish them“ (Samuels 1972, 65, invoking Bréal’s „Essai de semantique“ of 1897). The phenomenon of native-language synonym displacement has been documented amply in modern European languages. A case in point is traducere/to translate in the Romance vernaculars. In Medieval Latin translatare had been used, and had entered Romance languages; via French it was received into English, where it still survives. Traducere in this meaning was a coinage in Early Modern Latin of the early 15th century. It entered Romance languages in the late 15th century and replaced pre-existing words derived from translatare within a hundred years (cp. ital. tradurre, fr. traduir). Little attention has been paid to the fact that the situation in Latin itself is much less clear cut. Certainly no synonym displacement can be seen in Latin concerning the synonyms for to translate, neither in the Latin of Antiquity (i.e. in the Latin of native speakers) nor later (Ramminger 2015-16). Hence, some basic data about Latin synonyms may be useful.

I use the term ’synonym‘ as in the preliminary definition given by Wiegand, as words „used identically or similarly as far as content is concerned“ (Wiegand 1975/1999, 21). Given that a large part of Latin literature of Antiquity is lost, and oral Latin is obviously inaccessible, we need to keep in mind that there may be nuances that appear only fleetingly and that, for the lack of comparative material, can no longer be diagnosed by the lexicographer.

The following is based on data published in the Thesaurus Linguae Latinae. The TLL is well suited for research on synonymy in Latin. Since it is based on lemmatized and disambiguated data which are interpreted semantically, the necessary information is accessible (as opposed to databases); due to its format and size (this is not a pocket dictionary!), the space for presenting that kind of information is at hand (as opposed to other Latin dictionaries). Mostly, information on synonyms is contained in the paratexts prefacing the subgroups of a lemma; occasionally the TLL offers information in tabular form.

prodigium / monstrum /ostentum / portentum

The first comparison I want to discuss, is of four substantives meaning ’sign, omen‘, which to a large degree can be used interchangeably.

Table. 1: prodigium, TLL IX, 2 col. 1605 (Brenner)

Diagram 1: prodigium and synonyms (selected authors)

The diagram makes two types of synonym distribution visible. The first one, of those authors who use all synonyms to a significant degree, is exemplified by Cicero (with a preference for portentum); in the same group we have the Elder Pliny (with a preference for prodigium and ostentum), and Augustine (with equal usage of monstrum and prodigium; ostentum here is the loser in the competition). The other type of distribution privileges one synonym, and hardly uses the others. Its most extreme exponent is Livy with his preference for prodigium. This diagram also shows the preference for monstrum  in poetry (Virgil, Ovid, Seneca tragicus, Lucan; already noted in TLL monstrum, VIII, 1446, 72, with literature). As far as the displacement of synonyms is concerned, our diagram shows no sign of it: while individual authors may ignore some of the words, they are alive and well before and afterwards.

metuo – metus / timeo – timor

A different type of comparison is given in a table under metuo: a comparison of substantive and verb of two synonymous lexemes meaning ‚fear‘, metuo/metus and timeo/timor. What interests us here is the fact that it puts numbers to the distribution of synonymous lexemes of different word type.

Table 2: metuo, vol. VIII, 901-902 (Bulhart)

Diagram 2: comparison metuo – metus / timeo – timor (selected authors)

These numbers are full of surprises. First, subcorpora within an author’s oeuvre do not necessarily behave the same way; authors can have different preferences in different genres. We have two opposite examples in the two authors, where we have the numbers for subcorpora. Cicero does not have a unified usage; his letters behave differently (preference for timeo) from the other subcorpora (speeches: metuo). Seneca’s tragedies and prose works, on the other hand, show the same preferences in both subcorpora. Secondly, substantive and verb do not necessarily behave in the same way: Livy prefers metus to timor, but timeo to metuo, the same for Seneca; Tacitus prefers metus to timor, but has no favourite amongst the two verbs. Notable are the numbers for the Vulgate: Jerome has a marked preference for both timor and timeo. Unfortunately this is the only Christian text in our table; it would be interesting to know whether his choices are reflected in later Christian texts. But again, no sign of a permanent displacement of synonyms.

Thus, as far as synonym displacement is concerned, the question remains open. Like any living language Latin changed over the course of a millennium of written texts, and the data above are only a beginning. A desideratum would more Late Latin data. With the gradual dissolution of the linguistic unity of the Latin West the treatment of synonyms might have become more similar to the situation in modern European languages.

Obviously, there are limits to the interpretation of these numbers. Importantly they are not z-scored (adjusted to the text length); thus the individual authors cannot be directly compared to each other. Also there may be other synonyms which – while equally important for some authors – are not mentioned in these tables (e.g. pavor occurs 163 times in Livy and is in this author more important than timor, see TLL pavor, vol. X,1 838, Hickson).

I hope this post has wet your appetite for more. The TLL offers the data for many other comparisons: laevus/sinister, idciro/ideo, imber/nimbus/pluvia, immensus/inifinitus/ingens, pavidus/timidus …


Johann Ramminger, „Language Change in Humanist Latin: the case of traducere (to translate)“. In: Analecta Romana Instituti Danici 40-41 (2015-2016), 35-62. Download here.

M. L. Samuels, Linguistic Evolution with Special Reference to English, Cambridge, Cambridge University Press, 1972.

Herbert Ernst Wiegand, „Synonymy and its Role in Monolingual Lexicography“, In: Id., Semantics and Lexicography. Selected Studies (1976-1996), ed. Antje Immken and Werner Wolski, Tübingen, Max Niemeyer Verlag, 1999, 11-53 (translated from: Synonymie und ihre Bedeutung in der einsprachigen Lexikographie, 1975 [1976]).

Schreibe einen Kommentar

Deine E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert.

Diese Website verwendet Akismet, um Spam zu reduzieren. Erfahre mehr darüber, wie deine Kommentardaten verarbeitet werden.