Negative adjectival morphology in Latin.

3.1 Three varieties of negation

As is well known, Aristotle was the first philosopher that systematized the issue of negation: his proposals on this topic are still acknowledged as the basis for discussion. “In the beginning of Categories Aristotle distinguishes four varieties of opposition, three of which inherently involve negation: contradiction, contrariety, privation, and correlation” (Horn 2020: 7).^[5] It may be useful to remind the reader that two terms are contradictory when they are mutually exclusive and exhaustive: for instance, a number is either even or odd and there is no other possibility given (tertium non datur): i. e. parity and disparity cover all possible numbers.

Two terms are contrary when they are mutually exclusive but non-exhaustive because there is another possibility (tertium datur): for instance, an object cannot be big and small at the same time, but it can simultaneously be neither big nor small.

Usually, in the literature on this topic true negation is distinct from other logic modalities, which can, but do not necessarily have to, receive a negative interpretation since their content is not negative stricto sensu. Already Jespersen (1917) observed that the contradictory adjectives could develop a contrary meaning over time. The distinction that will be employed in the analysis presented in the following pages is based on the Aristotelian opposition that still remains crucial between contrary (ἐναντίος) and contradictory (ἀντιφατικός).

The third variety of opposition that is supposed to inherently involve negation is privation: as a type of negation, it is particularly challenging and it raises several peculiar problems. Its analysis requires a specific line of reasoning that does not take into account the logic relation between the negative entity and its positive counterpart, but rather the existence or non-existence of the entity in the scope of negation, as will be shown below. However, even in this case, privation is a multifaceted concept and in principle it could be seen as a hypernym of negation because any negative noun or adjective could probably be paraphrased as ‘without X’: unhappy ‘without happiness’, unconscious ‘without consciousness’, incorrectness ‘without correctness’, and so forth. However, this picture does not correspond to the complexity of language and such an automatic reformulation would end up concealing the actual mechanisms involved in encoding negation.

In addition, if one pays attention to the adjectival category, one will observe that there exist adjectives that are formally positive but semantically negative. A good example could be represented by the English adjective empty. Even in Latin there are representatives of both types, i. e. uacuus ‘empty’, formally positive, and inanis ‘empty’, whose etymology is far from being certain: its initial in- is likely to correspond to the negative prefix in- and consequently it would be a negative adjective even formally.

3.2 Negation as separation/absence

Negation expressed through adjectives and nouns had already been conceived by Aristotle as a form of absence, lack of the quality or content of the term involved. So the adjective improbus, for instance, qualifies one who lacks the quality of being morally sound, imberbis is a man that has no beard, and so on. However, even though these two negative adjectives seem to have the same status and express the same type of negation, they differ from one another crucially, as will be shown below, Section 5.

Although over time other philosophers and logicians, notably Kant (1992 [1763]), for instance, accepted Aristotle’s approach to the issue of negation, it must be stressed in any case that from a conceptual viewpoint notions like lack (of something), (de)privation, absence, even loss, do not perfectly overlap with that of negation. Apparently, there would be a conceptual difference between the true negation that manifests itself in the sentence at the syntactic level, and the morphological negation that has to be considered rather as a form of lack or absence. The result of this approach is that negation on nouns and adjectives is not a semantic primitive as one might take for granted. Looking at the etymology of the prefixes that convey the negative value in Latin, they can always be traced back to adverbs or particles, whose meaning or function refer to the semantic field of separation.^[6] There are parallels in other languages like Finnish, for instance, where a phrase with privative meaning is marked by the abessive case.

The fact that such prefixes have to do with the notion of separation, both etymologically and semantically, and that only secondarily they develop the value of negation, needs some reflections.

Since these prefixes are related to the notion of separation, they frequently also possess, or develop, a pejorative value over time, or even indicate an approximation. Therefore, adopting the notion of approximatives from Wierzbicka (1986, where the notion is only “positive”), they possibly should be described more adequately under the label of “negative approximatives”. The only prefix with an intrinsically negative meaning is in-, which goes back to Proto-Indo-European *n̥, whose development is recorded in almost all branches of Indo-European. However, even this prefix is supposed to be originally an element with a privative meaning (the Greek grammarians labelled it as ἀ στεϱητιϰόν); see Section 5 below.

3.3 Negation as privation

The label of privation itself is not completely unambiguous: there are at least two possible ways of conceptualizing privation, as already proposed by Kant 1992 1763: 782),^[7] either as a type of absence (Abwesenheit) or of lack (Mangel). According to the interpretation that is required by the context in which privation is expressed, there are some crucial linguistic differences and consequences.

If privation is conceptualized as Absence, then all adjectives that express negation by means of prefixes, whose primary meaning is ‘away from, separated from’, should be interpreted as adjectives conveying a meaning of privation, at least from the etymological viewpoint. In any case, it refers to something optional, i. e. that does not have to be necessarily present. This could be the case for which the label Privative can be employed.

On the other hand, if privation is conceptualized as a form of Lack, it means that something necessary is unexpectedly absent. In this case, the label that perfectly fits is Caritive.^[8] Both terms, i. e. Privative and Caritive, are quite commonly used in linguistic typology.

3.4 Aristotle’s ideas on privation

Interestingly, Aristotle put forward two possible interpretations of the concept of privation, albeit implicitly, without distinction of whether it was a type of absence or a type of lack. This is well illustrated in the following passage, taken from the Categories (12a28–33):

Στέϱησις δὲ καὶ ἕξις λέγεται μὲν πεϱὶ ταὐτόν τι, οἷον ἡ ὄψις καὶ ἡ τυφλότης πεϱὶ ὀφϑαλμόν· καϑόλου δὲ εἰπεῖν, ἐν ᾧ πέφυκεν ἡ ἕξις γίγνεσϑαι, πεϱὶ τοῦτο λέγεται ἑκάτεϱον αὐτῶν. ἐστεϱῆσϑαι δὲ τότε λέγομεν ἕκαστον τῶν τῆς ἕξεως δεκτικῶν, ὅταν ἐν ᾧ πέφυκεν ὑπάϱχειν καὶ ὅτε πέφυκεν ἔχειν μηδαμῶς ὑπάϱχῃ· νωδόν τε γὰϱ λέγομεν οὐ τὸ μὴ ἔχον ὀδόντας, καὶ τυφλὸν οὐ τὸ μὴ ἔχον ὄψιν ἀλλὰ τὸ μὴ ἔχον ὅτε πέφυκεν ἔχειν· τινὰ γὰϱ ἐκ γενετῆς οὔτε ὄψιν οὔτε ὀδόντας ἔχει, ἀλλ’ οὐ λέγεται νωδὰ οὐδὲ τυφλά

‘“Privatives” and “Positives” have reference to the same subject. Thus, sight and blindness have reference to the eye. It is a universal rule that each of a pair of opposites of this type has reference to that to which the particular “positive” is natural. We say that that is capable of some particular faculty or possession has suffered privation when the faculty or possession in question is in no way present in that in which, and at the time at which, it should naturally be present. We do not call that toothless which has not teeth, or that blind which has not sight, but rather that which has not teeth or sight at the time when by nature it should. For there are some creatures which from birth are without sight, or without teeth, but these are not called toothless or blind’ (Tr. Edghill).^[9]

In this quotation, Aristotle illustrates privation as a concept strictly tied with conditions set by nature. More precisely, he distinguishes between the condition of being without teeth in a period when teeth should be present by nature and the condition of being without them because teeth have not yet come through. The former case is that of old people, for instance, who have lost their teeth getting older, the latter is the case of newborns, who start cutting their first teeth when they are about five months old.

This distinction mirrors two situations crucially different: in the former case, the negation concerns or affects something that exists or used to exist, whereas in the latter what is negated does not exist or has not yet come into being. Quoting one of Aristotle’s examples, the Greek adjective νωδός^[10] ‘toothless’ always refers to somebody who has lost his teeth, not to a newborn, who is by natural condition without teeth. Aristotle connects these two cases with two different conditions imposed by nature.

3.5 Negation and presupposition

However, by slightly changing the perspective and equating the natural condition of being present with the presupposition of existence,^[11] wider typological generalizations are possible. For instance, there is a parallel to the distinction of presupposition just illustrated that concerns verbs: even in this case presupposition turns out to be crucial whenever the negative meaning is involved, as illustrated in Section 3.6 below.

Aristotle had already noticed that the negative prefix of Ancient Greek α- used to imply privation. Etymologically and also functionally, the privative α- perfectly corresponds to the prefix in- of Latin:

This passage belongs to a long section of Metaphysics (1022b3–1023a5), in which Aristotle claims that for all the types of privation there is a corresponding negative sense marked with the negative prefix ἀ-. Among the examples quoted there is one involving again a physical impairment such as blindness or defect such as being one-eyed. These examples are similar to the one involving the condition of being toothless mentioned in the Categories quoted above, in Section 3.4.

Aristotle’s passage deserves attention for several reasons, many of which are beyond the purpose of this paper and cannot be dealt with here. However, what is crucial here is the fact that a distinction is drawn between privation of something that is possessed at the right time according to the laws of nature such as teeth, and privation of something else, whose presence is not required by nature at any time.

3.6 Negation and existence

The distinction just mentioned recalls, and is parallel to, the logical difference between the so-called affected object and the effected object. Such parallelism, which has gone unnoticed thus far, consists of this: both in the case of the expression of the object and in the case of negation as privation, the label in question can be interpreted in two different, opposite ways. In the case of the object, it is presupposed as existent when it is affected, while it is presupposed as non-existent when it is effected.

Traditionally, there is a distinction between the accusativus effectus and the accusativus affectus. The former does not presuppose the existence of the entity, mainly an object, created or built by the verbal meaning, whereas the latter presupposes its existence. Two examples suffice to illustrate this point:

In this case, it must be also pointed out that in the corresponding negative sentences the conditions of existence or non-existence do not change if compared with positive sentences:

The same happens in the case of privation, because the thing that is lacking either implies the presupposition of existence or the presupposition of non-existence. Since there is no possible situation in between, this relation shares its distinctive property with that of contradiction. In both cases, what turns out to be crucial is the presupposition of existence of the entity negated or of the semantic role of patient expressed in accusative.

Since the conditions imposed by nature to which Aristotle refers in the passage of the Categories cited above and the presupposition of existence taken into account in the case of the object affected overlap, in the present contribution it is suggested to reformulate the conditions imposed by nature as conditions of existence. In other words, the two notions end up coinciding. Therefore, the proposal put forward here is to analyze the cases illustrated by Aristotle as examples in which negation operates on entities that are presupposed to be existent or non-existent. The feature of presupposition of existence thus becomes crucial.

3.7 Negation and morphological typology

There is a last point that deserves attention: the typological aspect of morphological negation. As already mentioned, negation encoded by morphology has not yet received the attention it deserves and a full description of the different strategies exploited to express nominal or adjectival negation is not yet available, the only contribution being Joshi (2020). However, no matter what framework is employed to describe the morphological negation on nouns and adjectives, some typological constraints must be taken into account when dealing with the occurrence of specific morphological mechanisms even though “no systematic typological studies have been made on derivational negation, and thus no generally valid cross-linguistic generalizations can be offered” (Miestamo 2017: 427). For instance, whether a language encodes negation morphologically to negate nouns and/or adjectives is a reasonable issue only in the case that the language in question structurally belongs to those in which nouns and adjectives are clearly distinguishable. If such distinction exists, the commonest strategy to negate adjectives or nouns among the world’s languages is probably by means of affixes.^[12] A sketchy typology of the morphological strategies of negation in nouns and adjectives is far beyond the purposes of the present pages, but some observations can be useful for illustrating the line of reasoning adopted here.

According to the typological structure of the language involved, negation is encoded differently. For some language families such Eskimo-Aleut or the languages of Native North America, the issue discussed here does not make any sense, because the parts of speech are not as distinct as they regularly are in the Indo-European languages. On the contrary, in a language belonging to the inflectional-fusive type, morphological negation can be encoded either as a prefix,^[13] as in the case of the Indo-European languages that still witness this phenomenon, including Latin, or as a suffix, in languages typologically different such as Turkish. In this language the suffix -sIz, literally ‘without’, corresponds to the privative, and transforms the nouns on which it is marked into negative adjectives: zamansιz ‘inappropriate’ (lit. ‘without time’), anlamsız ‘meaningless’ (lit. ‘without meaning’). This mechanism also occurs in many other languages, for instance in Australian languages (Dixon 2002: 84–86; for a general view, see Miestamo [2017: 427], where even more complex cases are mentioned), or in some European languages such as English and German, characterized by the suffixes -less/-los and -free/-frei. These suffixes, either in English or German, intrinsically presuppose the existence of the entity they negate, tend to be preferably associated with a concrete object rather than with abstract qualities. It is worth noting that the suffix -less as in toothless seems to be preferably interpreted as a privative, the suffix -free as, for instance, in alcohol-free, preferably as a caritive. Obviously, this point, for which also the notion of alienability or inalienability could play a role, requires further investigation.

4.1 General remarks

It is difficult to determine the precise number of prefixes in Latin that could convey a negative value on nouns and adjectives or change positive terms into their negative counterparts.^[14] There are seven prefixes, i. e. bound morphemes, here listed alphabetically, that perform this, even though to a different degree: dē-, dis-, ex-, in-, nĕ-, sē-, uē-. Each of them was also productive to a different degree and documented for different spans of time: the prefix in-, for instance, became the only negative prefix par excellence, whereas the prefix nĕ- disappeared.^[15] On the other hand, the prefixes nĕ- and sē- show a low degree of productivity and the number of negative adjectives they form is quite small indeed. Obviously, it would be very nice if we could ascribe a different semantic nuance to each prefix. Unfortunately, at least for the time being, this is a difficult task that remains beyond the purpose of the present contribution. Surely further investigations are needed.

It must also be pointed out that in Greek there are many negative adjectives that formally correspond to those of Latin, mainly but not exclusively, those that start with in-, so that it is not always easy to ascertain whether the Latin adjective is an independent creation or a calque. Even this point needs an accurate scrutiny of all adjectives recorded, which is not possible here.

In the Section 4, an overview of the Latin prefixes with negative value will be sketched, finally concentrating on the prefix in-.

What follows is a concise specimen showing how the prefixes mentioned above change the meaning of the adjectives with which they combine, from positive to negative.

4.2 Prefix dē-

dē-: “Komposita mit dē- ‘weg-’, ex- ‘heraus-’, vē- ‘weg-’ bezeichnen, ähnlich wie die mit in- priv., als gewissermassen negative Besitzkomposita den Verlust oder das Fehlen eines Besitzes” (Leumann 1972: 398). The derivational base seems to be exclusively nominal. The semantics of this adjective refers to the diminishing intensity of the quality involved. In other words, by means of the prefix dē- it is not the nominal content that is completely negated, but rather its fullness, so that a word like dēcolor does not mean ‘without color’ but ‘discolored’.

Dēcolor: ‘deprived of its natural color’; Cicero (Tusc. 2, 8, 6)^[16] translated the phrase χλωϱὸν αἷμα occurring in Sophocles’ Trachiniae (v. 1058) as decolorem sanguinem.

Dēcrepitus: ‘very old’ lit. ‘noiseless’ (Plaut. As. 863)

Dēdecorus: ‘disgraceful, shameless’ (Plaut. Bacch 1191)

Dēgener: ‘degenerate’ (Virgil, Aeneid 2, 549): “apparently a Virgilian coinage first found here” (Horsfall 2008: 414). In Servius’ commentary the adjective is explained as “non respondentem moribus patris”.

Some adjectives exhibit a different formation, in which the prefix is added to the stem:

Dēformis (forma): ‘unshapely’ (Cic. Inv. 1, 135).

Dēsomnis (somnus): ‘deprived of sleep’ (Petr. 47, 5): it is a hapax legomenon, possibly built after insomnis or exsomnis, both occurring in Virgil (Aen. 9, 167 and Aen. 6, 556, respectively) or, less probably, dēformis.

In the case of dēbilis ‘weak’ the stem belongs to a word that did not exist (any longer?) in Latin, etymologically connected with Skr. balám ‘strength’. dēbilis ‘weak’: literally ‘who has not his/her full strength’, as in Cato’s example (agr. 157, 10):

4.3 Prefix dis-

dis-: This is the definition offered by Ernout and Meillet: “Marque la séparation, l’écartement, la direction en sens opposés … et par suite le contraire, la négation …” (Ernout and Meillet 1985: 176; my emphasis).

The derivational base can be either adjectival or nominal. Even this prefix is usually associated with negation and, when derived with an adjectival base, is usually supposed to signal contrariety, not contradiction. However, the function exerted by this prefix raises some questions about this interpretation. Let’s take the example of the pair facilis ∼ difficilis: dis- does not really negate the content of the positive adjective, but rather rejects the viewpoint according to which something is easy to do and accepts the opposite viewpoint. The relationship between the two adjectives is reverse and is reminiscent of the pairs of relative antonyms (on this, see Lyons 1977: 280–282; Saeed 2016: 64). This seems to be explicitly stated in a famous verse of Terence:

This perspective also applies to the other adjectives:

Dispar: ‘dissimilar, unlike, different’ (∼ par). This adjective is attested from the age of Cicero in several authors, including Cicero himself and Lucretius.

Dissimilis: ‘dissimilar, different’ (∼similis). This adjective is also attested from the age of Cicero in various authors. It is worth noting that Plato (Parm. 129a1–3) used the similar pair ὅμοιος ∼ ἀνόμοιος to illustrate the relation of contrariety (ἐναντίον). However, it must be observed that dis- and ἀν- do not match perfectly.

Dissocialis: ‘antisocial’ (∼socialis). The fact that this adjective occurs in a late author like Rutilius Namatianus (1, 384) is good evidence that the productivity of dis- as a negative prefix lasted for a long time.

As mentioned above, there is a small group of adjectives derived from a nominal base and with a positive counterpart, usually with the prefix con-:

Discolor: ‘of another color’ (Varr. rust. 3, 17, 4) (∼concolor ‘of the same color’).

Discors: ‘discordant’ (Lucr. 5, 894) (∼concors ‘concordant, harmonious’).

Dissors: ‘of a different fate’ (Ov. am. 2, 12, 11) (∼consors ‘shared, in common’).

The relation between the adjectives with dis- and their counterparts with con- cannot be the same as in happy ∼ unhappy, but is closer to contradiction rather than to contrariety.

4.4 Prefix ex-

Ex- as negative prefix forms adjectives from a nominal base. Formally, in a few cases these adjectives exhibit a phonological readjustment as in ēlinguis. In a few cases they also exhibit the stem of both the adjectival inflectional classes as in ēlinguis/ēlinguus, even though the form belonging to the second class of adjectives is in general the older. The prefix ex- indicates the loss of something inherent and inalienable like some body parts, and explicitly expresses the caritive.

exanimis: ‘lifeless’ (Virg., Aen. 5, 517) /exanimus ‘lifeless, dead’ (Lucr. 6, 1256).

exanimalis: ‘lifeless’ (Plaut. Bacch. 848).

ēdentulus: ‘toothless’ (Plaut. Most. 275).

ēlinguis (Pacuv. 176)/elinguus: interestingly the grammarian Charisius (gramm. p. 399, 26) informs us how the word was interpreted: “elinguis habet linguam, sed usu eius caret; elinguatus amisit”.

expers (already in Plautus’ comedies; in Turpil. com. 157 it still occurs as expars): ‘having no part in, free from’.

exsensus (Laev.frg. 8)/exsensis: ‘senseless’, is an extremely rare adjective, as can be seen in the entry of TLL.

It must be observed that adjectives like ēdentulus or exanimalis are typical examples of the well-known linguistic creativity of Plautus and must be analyzed and evaluated cautiously, because they might not reflect the really spoken language but might simply be a new, funny coinage of the comedian.

4.5 Prefix nĕ-

nĕ-: this prefix is the full grade negation, whose zero-grade is represented by *n̥, which gave en- > in- in Latin. This prefix occurs with both nouns and adjectives, the majority of which belongs to the oldest layer of the language:

necessus: ‘indispensable’ (from Plautus onwards);

nefastus: ‘impious, irreligious’ (XII Tables in Cic. Leg., 2, 8).

4.6 Prefix sē-

sē- (sēd-, sō): Ernout and Meillet describe this prefix, with several allomorphs, as a “particule marquant la séparation, l’éloignement, la privation” (Ernout and Meillet 1985: 609). Cf. sēd, sē fraude ‘without fraud’ (XII Tables). According to de Vaan (2021: s.v.) “[t]he abl.sg. (* swēd? >) * sēd originally meant ‘per se, for itself’, whence ‘however, but’ as cj. and ‘without, away’ as prep.” It encodes the absence of an element and corresponds to the preposition sine ‘without’. The scantiness of instances makes it very likely that this prefix never was particularly productive. Whereas sēcūrus and sēdulō are the result of the lexification of a phrase, so that they do not have any positive counterpart, in the case of sōcors it is difficult to establish whether it derives from a phrase *sē corde or not:

sēcūrus: ‘careless’ (<*sē cūrā ‘without care’; from Cicero’s age onwards);

sēdulō: ‘diligently’ (<*sē(d) dolō ‘without fraud, deceit’). This adverb, from which the adjective sedulus ‘diligent’ was derived, is attested from Plautus onwards;

sōcors: ‘slack, inert’ (from Plautus onward).

4.7 Prefix uē-

uē-: This prefix is difficult to etymologize and several proposals have been put forward. In Ernout and Meillet, uē- is described as a “particule privative ou péjorative … Ce préverbe figure au premier terme des composés à valeur négative” (Ernout and Meillet 1985: 716; my emphasis).

According to the piece of information provided by Paulus Festus’ text cited here, the prefix uē-, which was no longer productive during the classical age, conveyed a specific meaning, i. e. privation as lack of achievement:

In this example, uegrande frumentum refers to the corn, whose ears did not fully grow. The value encoded by uē- can be paraphrased as ‘almost getting to the point of being fully X’; it is synonymous with dē, with which it is in a relation of reverse (see Section 4.2 above). The majority of examples show that their bases were adjectival.

uēcors: ‘frenzied’. According to Fest. 373, 12; est turbati et mali cordis. From Pacuvius (trag. 216) onwards.

uēgrandis: ‘not very large’ (Plaut. Cist. 378).

uēsanus: ‘not of sound mind’ (Plaut. Trin. 826).

In Latin, another prefix, homophonous with uē- has been hypothesized, with the function of superlative as in uēpallidus ‘very pale’, a hapax legomenon occurring in Horace, sat. 1, 2, 129, but its reading is far from being certain and already the grammarian Porphyrio preferred to read vae pallida. However, according to the hypothesis accepted in the etymological dictionary of Latin by de Vaan (2021: s.v.), “Reinterpretation of vē- as meaning ‘abnormal’ could then explain the formation of vē-grandis, vēscus and vē-pallidus (Hor.).”

5.1 General remarks

Among the prefixes that convey a negative value in Latin, in- plays a special role:^[17] it is the only one that can express all the varieties of negation under scrutiny, i. e. contrariety, contradiction and privation. This remark is parallel to what Aristotle says about the privative ἀ- (see Section 3.5).

It is easy to observe that, whenever contradiction or contrariety is involved, what is central is the relation of exclusivity and exhaustivity between the two terms in question, whereas privation is involved in the existence or non-existence of the abstract or concrete entity ascribed to the negative term, i. e. adjective or noun. In Section 3.2 above, the pair improbus and imberb i s has been mentioned as an example of negative adjectives with the prefix in-. However, the two adjectives, albeit formally identical, are crucially different because of their semantics. A reformulation as ‘without probity’ and ‘without beard’, although basically correct, would not catch any crucial difference; still, a crucial difference between them exists. In the pair probus ∼ improbus there is a relation of contrariety. However, I would prefer to describe it as a case of neutralization, as will be clear below. As a matter of fact, in cases such as this one the relation between the two adjectives could possibly be described more adequately as a relation of neutralization of the quality involved, the terms contrary/contrariety rather suggesting an opposition.

In the case represented by the pair barbatus ∼ imberb i s, the relation between the two adjectives is different in several respects:

1. The pair does not conform to the pattern X ∼ in-X as represented for instance by pairs like probus ∼ in-probus (= improbus, after the phonological readjustment) or felix ‘happy’ (originally ‘fertile’) ∼ infelix ‘unhappy’.

2. The positive member of the pair is frequently represented by a past participle, whose meaning is ‘provided with X, endowed with X’.

3. The corollary of point 2 is that an adjective that was originally a past participle does not necessarily have a negative counterpart in an adjectival pair: for instance, beatus ‘blessed, happy’ does not have any corresponding counterpart *imbeatus ‘unblessed’.^[18]

4. The variety of negation is privation and the entity involved does not refer to an abstract quality but to a concrete element like the beard in the pair barbatus ∼ imberb i s.

5. The decisive feature is the necessary existence or non-existence of the entity involved: barbatus means ‘endowed with beard’, and the beard is taken as a physical element that necessarily accompanies the growth and development of the male body. The adjective imberb i s (or its variant imberb u s) presupposes the natural absence of the beard. It corresponds to Eng. beardless, It. imberbe, and can never be translated as shaved or It. sbarbato because this adjective, an original past participle, presupposes the existence of a beard to shave. Thus imberbus would rather mean something like ‘one, who is without beard because it still has to grow’ and not ‘one, who has shaved his beard’, by implying that the condition of ‘necessity to be’ is absent from ‘beardless’.

5.2 The prefix in- with adjectives

As shown, the negative adjectives with the prefix in- can be formally subdivided into three groups, the last of which can be further subdivided into two subgroups:

Group A: contains adjectives derived from adjectival bases; the prefix leaves formally unchanged the adjectives to which it applies or shows phonological readjustments. The pair could be represented as X ∼ in + X like the type felix ∼ infelix. This is the case in which the relation between the two members may be more accurately described as a form of neutralization of the quality involved instead of a real negation.

Group B contains adjectives that do not have a positive corresponding form, but only a nominal base from which they are derived: the pair could be represented as the type: X_N ∼ in + stem of X_N + ending like bellum ∼ in-bell-is.

It is worth observing that Priscianus (gramm. II 9, 25) informs us about the way the adjective informis could be interpreted in the 6th century: “dicitur mulier, non quae caret forma, sed quae male est formata”. The original meaning of informis ‘that has no form’ was no longer in use and the adjective only meant ‘unshapely, deformed, hideous’.

5.3 The prefix in- with verbs

Group C needs some remarks. Its peculiar feature is that all these adjectives derive from the verbal base both of the infectum and the perfectum. The analysis of this group of adjectives was thoroughly accomplished by Arias Abellán (1996) in a paper devoted to the comedies of Plautus. She recognized two cases, here indicated as C1 and C2:

In this case the adjectives mean ‘one who does not know’, ‘one who does not contain’, and so on. Obviously, all adjectives develop meanings that are far from the original ones. For instance, innocens, originally ‘one who does not do harms, injuries’, developed the more common meaning ‘innocent, naif’.

The subgroup C2 can be subdivided into two other subgroups: there is a difference concerning the negative adjectives derived from the past participle, and it has gone unnoticed so far, at least to the best of my knowledge. According to the semantics of the verb, the scope of negation is different. Usually, scope is a notion employed in the analysis of the syntactic negation, but it also turns out to be necessary with this group of adjectives.

C2A. When the verb expresses an activity and therefore does not possess any resultative meaning, the scope of the prefix in- is over the entire activity. See for instance this example:

In this example, incenati means not only that they did not even start eating their dinner, but also that they did not receive any dinner. In the case of incenatus, since there is no other situation possible, the relation between cenatus and incenatus is of contradiction.

C2B. However, when the verb expresses a resultative meaning, the scope is exclusively over the result, not over the activity expressed by the verb. For instance, incoctus does not properly mean ‘raw’, but ‘not completely, perfectly cooked, undercooked’, as is clearly shown in the following instance:

In cases as the one illustrated by the verb coquere, it turns out to be difficult to decide whether the relation between the two forms of the pair is contrariety or contradiction. However, since the meaning is gradable, and therefore scalar, it has to be interpreted as a case of contrariety. In this case “the idea that the relevant distinction to work with in the domain of negation is the distinction between low and wide scopal negative markers rather than between morphological and syntactic negation” (De Clercq 2021: 12) turns out to function perfectly.

For the time being, the conclusions that can be drawn are that:

1. whenever the scope of negation is on the stem, as in C2A: IN + STEM + to-,^[20] the relation between the two members of the pair is of contradiction;

2. whenever it is on the notion of accomplishment or telicity, as in C2B: IN + stem + TO-, expressed by the verbal form usually coded by the past participle, the relation is of contrariety.