A key term in the sex/gender system informing the Latin textual tradition is the adjective mollis (‘soft’) along with its adverb molliter, the abstract noun mollitia or mollities, and the verb mollire. When referring to human beings, this set of words is closely associated with femininity, and the attribution of mollitia to men assigns a range of qualities to them which are marked in Roman terms as woman-like: physical attributes of softness and tenderness; behaviors or characteristics such as timidity, self-indulgence, or lack of self-control; an erotic style marked as feminine but in connection with female sexual partners (the “womanish womanizer” is a distinct stereotype in Latin literature); and the specific pleasure of being sexually penetrated. In what follows, rather than aiming for a thorough lexical analysis, I highlight some aspects of the semantics of mollitia which are most relevant to the sex/gender system, and suggest in a selectively illustrative manner some of the ways in which the lexicographical resources long available to classical philologists can be supplemented by tools which have been developed by linguists working in the field of lexical semantics. I limit myself in this essay to mollis and related words, but what follows also applies to any and all relevant lexical items in the Latin language of sex, gender, and sexuality1.
Lexical semantics: Introductory remarks
Lexical semantics can be defined as the study of meaning in individual words (as opposed to meaning at the level of sentence, utterance, or social or performance contexts) in natural languages (as opposed to artificial or computer languages). There is by no means consensus on the larger theoretical issues which have been debated for centuries and in some cases millennia: how to describe the relationships between words and concepts, language and experience. Nor is there unanimity on the relevant terms and concepts. Far from it: on nearly every point I mention below, linguists have disagreed and continue to disagree on terminology and conceptualization; every introductory text on semantics organizes and describes its subject in its own way; and nearly every example or case study cited in studies of semantics is up for debate2.
There are, to be sure, some significant qualifications on the extent to which we can apply the theoretical models and terminology of lexical semantics to ancient Greek or Latin. Most banal, but hardly insignificant, is the fact that we have only limited access to the full original range of word usage across all the “secondary” or written speech genres in these languages, and little or no access to the entire range of “primary” speech genres of direct, interpersonal, verbal communication3. More significantly still, key arguments about semantics are generally supported by appeals to judgments about which usages are normal, acceptable, questionable, or simply impossible, judgments which are based on the perceptions and intuitions of native speakers as they emerge, for example, from questionnaires and linguistic experiments, or from reflections by linguists themselves on their own native language. None of this, of course, is possible in the case of languages which have no native speakers. Yet Greek and Latin were, after all, natural languages and we can expect their semantics, like their morphology and phonology, to behave grosso modo just as they do in any other natural language. The attempt can be made, as long as we remain conscious of the limitations confronting us and as long as we are cautious about making assertions regarding the “normal” or “acceptable” or “standard” senses of a given word, or about what might be “questionable” or “impossible” Latin.
Although there is no comprehensive system of terminology universally adopted by specialists in lexical semantics, there is widespread consensus on some basic points. It is generally agreed, first of all, that discussion will always be incomplete and ongoing, not only because of the practical consideration that disagreement is possible on nearly any illustration or model proposed, but also because of the larger point that word meanings and distinctions between them are by their very nature characterized by a high degree of indeterminacy or “fuzziness”4. Most linguists, moreover, distinguish between words and lexemes, the latter defined as “abstractions of actual words that occur in real language use”, parallel to phonemes (abstract representations of linguistic sounds), as “units of the lexicon” or “vocabulary-units of a language”5. Love, loved, loving are three different word-forms but one and the same lexeme; bat (‘flying animal’) and bat (‘instrument for hitting a ball’) are the same word-form but two different lexemes6. Likewise it is generally agreed that the very term meaning is insufficiently precise: introductory books on semantics written in English regularly begin with a demonstration of the multiple “meanings” of the verb to mean and the noun meaning. Although there is no universally accepted system for describing the different ways in which a lexeme can be said to have meaning, one tripartite scheme is common. Denotation is the class of entities of which a proposition using a given lexeme is or is not true, and has to do with its relationship to extralinguistic categories (e.g. dog as an animal with various characteristics which distinguish it from cats, pigs, or birds); a lexeme’s sense is one specific way in which it is used, as opposed to other ways in which the same lexeme is used, a matter of a lexeme’s relationship to other lexemes (e.g. dog as opposed to hound or bitch or puppy or in relation to such lexical expressions as to dog, dog tired, dog-eared); and a lexeme’s reference is the specific entity at stake in a particular utterance (e.g. this dog as opposed to the class of dogs in general). In what follows, then, I am interested in various senses of what I call the moll- lexeme, and among the wide range of questions relevant to semantics, I focus on only three: meaning as detectable in usage patterns as catalogued in standard lexica; the relationships between certain senses of moll- and relevant senses of other lexemes (sense-relations); and the relationships among various senses of the moll- lexeme itself (polysemy).
Finding Meaning in Usage
If, then, there is one thing that the theorizing and practice of linguists working in the area of semantics have taught us, it is a lively skepticism regarding simple statements about what a given word “means”. But a good place to start is a review of usage, and Latinists are fortunate to be able to draw on a long tradition of lexicography, most richly embodied in those volumes of the TLL which have been published to date. The articles of this lexicon catalogue usage in dazzling, sometimes dizzying detail, and their carefully articulated structure allows readers with sufficient Latin to quickly perceive some important distinctions: when, for example, the adjective mollis refers to human beings as opposed to animals, things, qualities, or actions; whether it is being used in its “proper” or “transferred” sense (proprie or translate); whether the sense has a positive or negative value (in bonam partem or in malam [sometimes deteriorem] partem). And there are frequent indications of other Latin adjectives with which mollis has a relation of similarity or synonymy (i.q.) or contrast (opp.). See Figure 1 for the structure of the TLL article on the adjective mollis.
But precisely because of their mass of detail, intricate outline structure, and exclusive use of a spare Latin, TLL articles – particularly such long and complex articles as that for mollis – do not make it easy for their readers to perceive recurring themes or internal connections among the senses of a given word, or between a given word and others, whether related (mollis and mollitia) or not (mollis and muliebris). Even on its own terms, the terminological apparatus of the TLL raises without answering some basic questions of semantics. Are we to take i.q. literally, “the same as”? Surely not; but more precise distinctions of synonymy or near-synonymy are left unmade. Where is the line drawn between i.q. and fere i.q. and why? While the distinction between in bonam and in malam partem is of obvious utility, its either/or quality necessarily flattens out cases of ambiguity or questions of degree, and raises without being able to answer the question of why, and with reference to which cultural norms, certain senses are “good” or “bad”. The more compact and user-friendly articles of the OLD, for their part, allow readers quickly to perceive basic features of a word’s semantics. See Figure 2 for the structure of the OLD article on the adjective mollis. But OLD articles share with those of TLL a tendency to make such a large number of distinctions (in this case, on the basis of the kinds of things, qualities, or persons of whom the given word is predicated) that one finds a small number of English glosses (e.g. “soft”, “gentle”) scattered throughout the article, raising the question of whether the various subentries indeed correspond to what semanticists would describe as different senses of the lexeme.
Lexicographical practice fragments and potentially obscures in another way, as an inevitable result of the traditional practice of creating separate articles for adjectives (with or without adverbs), nouns, and verbs. Moreover, neither TLL nor OLD articles easily enable an attempt at a diachronic narrative: it is hard to ask of this material as presented whether there are detectable shifts in the semantics of mollitia between Plautus and Apuleius7. And while it is not, to be sure, the function of dictionary entries to construct narratives about the larger conceptual and cultural habits embodied by and perpetuated in language (what might be the linguistic or cultural logic whereby one and the same adjective can refer to poems on the one hand, and men’s preferred sexual partners on the other?), the TLL and OLD articles on mollis illustrate how lexicographical practice of this kind can sometimes hinder attempts to create such narratives. Something falls between the cracks when Ovid’s use of the language of mollitia to describe what happens to Hermaphroditus’ body (it is melded with that of the nymph Salmacis and his masculinity is thereby compromised) appears in the TLL as an illustration of the “proper” sense of the verb mollire – ‘to soften’ in the sense ‘to relax the firmness or stiffness of bodies’ – tucked away under the subheading “in other ways” (as opposed to softening by crushing, heating, or moistening)8. It is striking, if not positively confusing, that at mollis I.A.1.c.i.1, although the heading speaks of boys and women (fere mulierum vel puerorum), the first passage cited refers to neither a woman nor a boy but instead plays with the intimate association of the lexemes moll- and cinaed- (Plaut. Aul. 422: ita fustibus sum mollior magis quam ullus cinaedus).
In short, one of the questions that has been driving scholarship on sex and gender in the Latin textual tradition is to explore precisely what might be meant by glosses like effeminatus and effeminate, scattered liberally throughout the TLL and OLD articles, and to interrogate the equivalence these articles seem to establish between effeminatus, dissolutus and iners, effeminatio and infirmitas, between “effeminacy” and “weakness of mind or character”, “effeminately” and “erotically”9. In order to do so, we need to pose new kinds of questions of the material so painstakingly catalogued in the lexica. For example, reading the material in the TLL and OLD articles on mollis and related words, carefully divided into types of things or people who are ‘soft’, we might ask what happens if draw a distinction on the point of desire: ‘softness’ as attribute of subjects as opposed to objects of desire. One thing we see if we do so is that ‘softness’ is a quality valued in objects of desire, whether the bipartite group “women and boys” traditionally configured as attractive to men, or males (both boys and men) configured as attractive to women10. When, on the other hand, we consider the attribution of ‘softness’ to subjects of desire, we find a significant distinction: as a quality of female subjects of desire mollitia seems normative11, but as a quality attributed to male subjects of desire it is negatively valorized. The language of mollitia describes the stereotypical “womanish womanizer” who aims to attract women precisely by his “soft” and “feminine” manner of walking, for which he is criticized12, and it refers to a type of feminine man who enjoys being penetrated13. I have argued elsewhere that this mollitia neither denotes nor connotes either homosexuality or heterosexuality. To be sure, some of the “soft men” (molles) of Juvenal’s satires are imagined as seeking sexual pleasures with men (Juv. 2.47 magna inter molles concordia; 2.164-5 Armenius Zalaces cunctis narratur ephebis / mollior ardenti sese indulsisse tribuno), yet in Phaedrus’ poetic fable the feminine men who enjoy being penetrated (molles mares) and the masculine women who enjoy penetrating (tribades) are ideal partners for each other14. And then there are those perfumed dandies, sashaying about in order to appeal to women and, perhaps even more interesting, the women who are turned on by such “soft” and womanly men (Sen. Contr. 2.1.6: incedentem ut feminis placeat femina mollius). In short, even a quick review of the usage patterns of mollis and related words as catalogued in standard lexica, supplemented by a reorganization of the material along the axis of subject vs. object of desire, reveals some of the complexities of desire and gender in the Latin textual tradition.
Sense-relations
As we have seen, TLL articles are sprinkled with signs such as i.q. or opp. which point to other lexemes with related or contrasted senses. Work done in the field of lexical semantics can help us take some first steps in a sustained, systematic interpretation of these signposts. For a major question posed by lexical semantics is how to describe with precision and accuracy the semantic relationships between lexemes, or more specifically between certain senses of different lexemes: for example between hot in the sense ‘having a high temperature’ (and not the sense ‘heavily spiced’) and cold in the sense ‘having a low temperature’. There is much debate and no universally adopted set of terminology, but one distinction is common. If the lexemes whose sense are being studied belong to the same word-class or part of speech and can, without violating syntactic rules, be substituted for each other within an utterance, linguists speak of paradigmatic sense-relations (such as those among the relevant senses of the adjectives hot, warm, cold, sunny, cloudy, all of which may modify the noun day). If they belong to different classes or parts of speech, linguists speak of syntagmatic sense-relations (such as those between the relevant senses of the adjective hot and of nouns like day, sauce, or head).
Paradigmatic semantic relations of closeness, nearness, or similarity are the realm of synonymy, and opposition, contrast, or exclusion are types of antonymy15. These terms are familiar enough, but linguists have explored precisely how they work, and sometimes come to conclusions rather different from common understandings. Semanticists are in general agreement, for example, that very few words in natural languages are absolute synonyms, i.e. totally substitutable in denotation, connotation and social meaning as well as in all syntagmatic relations: apparently absolute synonyms like funny and comical, big and large, bathroom and toilet and john, all differ in one or more of these ways. Much more common are near-synonyms; indeed, some argue that absolute synonymy hardly ever exists in natural languages16. As for antonyms, a distinction can be made between complementary antonyms, which seem to leave no room for middle ground (even and odd, dead and alive, to stay and to go), and contrary antonyms (such as short and tall, old and young, to love and to hate) where there is an obvious middle ground, and the lexemes are said to have a “scalar” or “gradable” quality. It has been observed, however, that even such complementary antonyms as dead and alive can take on gradable qualities (“He’s more dead than alive”) and it has been suggested that all complementary antonym pairs are potentially gradable. This obviously has ramifications that go beyond the narrowly linguistic. In the case of the Latin lexicon, for example, we can say that mollis stands in an antonymous relationship not only with the obvious durus but also with fortis (which itself, as we will see momentarily, has a close relationship with masculinity). See Figure 3.1 for a sample of texts illustrating these antonym relations. But are we dealing with complementary or contrary antonyms? The latter seems more likely and the pairs mollis and durus, mollis and fortis, would seem to be gradable or scalar, a semantic point which fits a picture we find in the Latin textual tradition: the men of our texts are represented as not necessarily being entirely masculine or entirely feminine. Figures like Maecenas and Hortensius remind us that there is room for play and ambivalence, that there is a middle ground17.
Turning to syntagmatic semantic relations: if two or more lexemes belonging to different classes attract each other, they can be called philonyms (merry and Christmas), whereas if they result in a semantic clash, they can be called xenonyms (merry and Easter)18. Thus we might say that philonyms of the adjective mollis include the nouns puer and cinaedus as well as proper names of some Eastern peoples; that a xenonym of mollis is the noun vir; and that a philonym of vir in turn is the adjective fortis. See Figures 3.2, 3.3, and 3.4 for samples of texts illustrating these sense-relations. If we add these considerations to the point that certain senses of mollis are in an antonym relation with certain senses of both durus and fortis, we are reformulating in terms of lexical semantics some of the basic features of the sex/gender system informing the Latin textual tradition: that it is women and boys (rather than women and men, or females and males) who are identified as normative objects of male desire; and that women and boys are contrasted with fully masculine men – and therefore, it seems, also desirable – because of their qualities of ‘softness’ and ‘weakness’, which themselves stand in contrast with characteristic features of Roman masculinity. The attempt to describe the semantics in this way makes the binarisms underlying the system that much clearer, just as it reminds us that not all of the scenarios of desire which we find in the surviving textual tradition always fit these or other binary schemes.
Polysemy
It is a widespread perception among speakers of a given language that many (perhaps most, maybe all) words in that language have more than one “meaning”. The task of lexical semantics is to study the phenomenon in a systematic way, with special attention to recurring types and processes as a characteristic not only of lexical systems in themselves but also, in the wake of developments in the field of cognitive linguistics, as reflections of human cognition and culture19. Here, too, most linguists begin with a fundamental distinction. When differing senses of a single word-form are so different from each other that speakers of a natural language (sometimes in line with historical etymological differences, sometimes not) generally perceive them to be distinct lexemes, we are dealing with homonymy, and standard notation both in lexicography and in semantic studies is to number the lexemes with subscripts and to separate them out as distinct “entries” in the “mental lexicon”: bat1 as ‘flying animal’ and bat2 as ‘instrument for hitting a ball’; sole1 as ‘bottom of foot’ and sole2 as ‘type of fish’ (and in some analyses also the adjective sole3 as ‘alone, only’). When, however, speakers of a language generally perceive the existence of a single word with different senses, linguists speak of polysemy, and lexicographers print them as subheadings within a single dictionary entry: for example drink (1) ‘take in liquid through the mouth’ (“Dogs drink frequently”) and drink (2) ‘to take in alcohol-containing liquids through the mouth’ (“He only drinks at parties”). The fact that systems of numbering can be used to represent both phenomena – different lexemes with the same form (homonymy) and different senses of the same lexeme (polysemy) – points to the subject of ongoing debate. Where do we draw the line between polysemy and homonymy, and what justifies our decision? Native speakers’ perceptions do not always align with the historical facts of etymology; to which of these should we give more weight, and why? Or should we perhaps not even attempt to make a firm distinction between homonymy and polysemy, but instead look for a continuum20?
There is little agreement among linguists and lexicographers on this theoretical point, or indeed on the practical questions of how to distinguish among senses and lexemes, or how to describe in metalanguage or represent visually the relationship among the various senses. But one distinction is commonly drawn: that between vagueness or indeterminacy (the lack of specificity of a given sense with regard to one or more criteria) and ambiguity (the existence of multiple senses, i.e. polysemy proper). Various tests have been developed for establishing whether vagueness or ambiguity exists, and while there have been challenges to the validity of some or all of these tests, many continue to find value in the usefulness of making the distinction. And an important point is that “since vagueness, polysemy and homonymy apply at different levels of consideration (single sense, single lexeme, different lexemes), it is sometimes the case that a single word form can illustrate all of these phenomena”21. Thus we could say that the moll- lexeme is polysemous (with senses ranging from ‘soft’ to ‘cowardly’ to ‘womanizing’ to ‘enjoying being penetrated’), but that some of these senses are vague. Is a man who is mollis in the sense ‘enjoys being penetrated’ necessarily associated with male sexual partners? Elsewhere I have argued that such ‘soft’ men might be imagined as being penetrated by female partners; in other words, this particular sense of the polysemous lexeme moll- is vague with regard to the criterion of the sex of a man’s sexual partner22.
How might we visually represent the various senses of a lexeme and the relationships among them? One technique is to construct a polysemy network, sometimes assigning numbers to the various senses, often using arrows or lines to connect them, and arranging them on the page in various ways: radiating out from a core concept, for example, or as a tree23. Such networks complement traditional dictionary entries, not only because networks of this kind suggest a narrative of logical relationships between senses in a way that lists or elaborate outlines cannot, but also because they condense and streamline, with all the obvious advantages and disadvantages24. See Figure 4.1 for a first attempt at a partial polysemy network of the moll- lexeme, limited to those senses which are relevant to human beings and their gendering, and Figure 4.2 for some exemplary texts, taken from the TLL article, illustrating each sense.
This is to be understood as part of a much larger and more intricate network; its selective, streamlined nature is meant to draw attention to some fundamental semantic processes at work in the move from sex to gender. We see, for example, narrowing (from physical qualities associated with femaleness to the specific trait of being penetrated) and highlighting (emphasizing the element of sexual desire). We also see some metonymic associations: between physical softness on the one hand, and qualities like delicacy, softness, tenderness or inconstancy, self-indulgence, lack of self-control on the other; between possession of the penis and/or testicles and masculinity, such that their removal or incapacity signifies the absence masculinity; between performing a physical act and experiencing pleasure in the act. Especially significant is the metonymic shift visually represented at the head of this partial network: from sense 1 ‘soft’ to sense 2 ‘woman-like’, and thence to all other senses; the positioning of sense 2 highlights the point that all other senses are to one degree or another gendered as feminine. This network also allows us to see that sense 3 ‘cowardly, timid, etc.’ is fairly distant from the others, and the contiguity of the complementary senses 8 and 9 highlights the key role of sexual penetration in the Latin semantics of gender. Indeed, the semantic relations illustrated by this partial network draw attention to the association of women and the feminine not, say, with women’s status as givers of life, but rather with the sexual role of being penetrated.
A question with important implications is where to locate senses 6 and 7 (‘sexually attractive to women’ and ‘sexually attractive to men’), and whether to separate the two out or else collapse them into a single sense ‘sexually attractive’. I have tentatively distinguished between the two and placed them underneath sense 5 (‘indulgent in sexual pleasure, pleasure-driven, sensual’), thereby highlighting those features as that which arouses desire. But one conceivable alternative is to place them underneath sense 2, parallel to senses 3, 4, 8 and 9, thereby highlighting the feature of ‘womanish softness’ as that which arouses desires – crucially, in both men and women. To be sure, the positioning of sense 2 in this network reminds us that the language of mollitia when applied to human beings is always already gendered as feminine. But this alternative configuration would more prominently highlight the homoerotic configuration of one kind of feminine desire described in the Latin textual tradition: women are understood as potentially finding womanish men attractive.
An example of polysemy at work: A reading of Catullus 16
Many utterances come with cues for the selection or exclusion of specific senses: the sentence He only drinks at parties activates the sense drink (2) (‘consume alcoholic beverages’) by means of of the qualifications at parties and only. In other utterances, however, certain cues or their absence make it possible to create and perceive meaning precisely in the tension between senses. This is the realm of puns and jokes, deliberate or productive ambiguity, “double meanings” as they are called. In this final section of my paper I give illustrations of how the exercise of identifying such cues in texts which use the polysemous moll- lexeme can add further nuance to our readings of how gendering works in Latin literature.
In Phaedrus’ poetic fable on the origin of molles mares and tribades, cues for the selection of sense moll- (9), ‘enjoys being penetrated’ come in the form of the reference to “the maiden’s organ” (virginale) which a drunken Prometheus attaches to male figures who are called molles mares, together with the reference in the final line to the desire and pleasure these males experience, qualified as “perverse” (ita nunc libido pravo fruitur gaudio, 4.16.14). In Juvenal’s ninth satire, cues for the same sense of the moll- lexeme are found in the proximity of the phrase mollis avarus (9.38) to references to a money-conscious patron who “wiggles his buttocks” as he goes over his accounts (computat et cevet, 9.40), along with the climactic pronouncement that the addressee will always find a sexually receptive patron in Rome (ne trepida, numquam pathicus tibi derit amicus, 9.130). Arellius Fuscus’ scornful image of the daintily walking womanish womanizer underscores, by repetition and parallelism, the point that such a man aims to please women precisely in his womanly softness (ut feminis placeat femina mollius, quoted at Sen. Contr. 2.1.6), thereby pointing to sense moll- (7) ‘womanish man who is attractive to women; womanish womanizer’. With a differently structured repetition, a passage from Ovid’s Ars amatoria cues the same sense by urging a man with “soft arms” to use his advantage in order to please the woman he wishes to seduce (si mollia bracchia, salta, / et quacumque potes dote placere, place). And the blunt language of Juvenal’s sixth satire contains cues for the same sense of the moll- lexeme, naming two women and evoking the corporeal signs of their sexual pleasure as they watch a male dancer described as mollis (6.63-4: chironomon Ledan molli saltante Bathyllo / Tuccia vesicae non imperat, Apula gannit).
I end with a brief reading of a poem well known to students of sex and gender in the Latin textual tradition25. Catullus 16 begins by juxtaposing the brutal language of sexual penetration with the lexemes pathicus and cinaedus: pedicabo ego vos et irrumabo, / Aureli pathice et cinaede Furi (1-2). When the language of mollitia appears two lines later, and in juxtaposition with a negation of pudicitia at that (3-4: qui me ex versiculis meis putastis, / quod sunt molliculi, parum pudicum), the cues for the sense moll- (9) ‘enjoying being penetrated’ are clear. Yet as the poem, building up to its climactic repetition of the opening line, makes cross-reference to other Catullan poetry on kisses (vos, quod milia multa basiorum / legistis, male me marem putatis? 12-13), the cues now encourage the selection of other senses of the moll- lexeme. For not only is there no obvious link between “many thousands of kisses” and taking pleasure in being penetrated but, more tellingly still, the internal cross-reference effected by the verb legistis points us to other poems in which the Catullan persona is a man who kisses, desires, and implicitly desires to penetrate, both women and boys: in particular to 5.1-9 (vivamus, mea Lesbia, atque amemus. / ... / da mi basia mille, deinde centum, / dein mille altera, dein secunda centum, / deinde usque altera mille, deinde centum; cf. 7.1-2: quaeris quot mihi basiationes / tuae, Lesbia, sint satis superque); and to 48.1-4 (mellitos oculos tuos, Iuventi, / si quis me sinat usque basiare, / usque ad milia basiem trecenta / nec numquam videar satur futurus). To paraphrase: “You think me not fully a man because you have read my kiss poems? I’ll show you who’s the man”. The causal conjunction “because” (quod...legistis) leading to the internal cross-reference is of particular interest: they think him unmanly not because he has, for example, shown pleasure in being penetrated, but because he has so exuberantly written of so many kisses, with both Lesbia and Juventius. Not the objects of his desire but the style of his desire is at stake: not moll- (9) but moll- (6). Yet the poem’s parting shot, its repetition of the opening threat of sexual penetration, arguably responds to the charge of mollitia by suggesting all its relevant senses, perhaps particularly moll- (4), (6), and (8). In fact, one can read the final lines as vividly expressing a rejection of mollitia in any and all senses by making what in Roman terms is arguably the most fundamental gesture of masculinity: playing the penetrative role (12-14: vos quod milia multa basiorum / legistis, male me marem putatis? / pedicabo ego vos et irrumabo). This final, triumphantly masculine stance is made all the more vivid by the specification of two modes of penetration, anal and oral.
The intertwined processes of considering which senses of a polysemous lexeme are possibly or necessarily activated in any given use, which in turn means identifying what kind of cues can activate which senses and teasing out the relationships between the various senses, lie at the heart of our processes of interpreting texts. For us as non-native speakers reading Latin and ancient Greek texts which speak the language of sex and gender, the process is that much more conscious and that much more challenging.
* * *
FIGURE 1: Structure of the TLL article
| I. proprie pertinet ad corpora non dura neque rigida | ||||||
| A. strictius, i.q. tactui cedens, non durus | ||||||
| 1. Generatim | ||||||
| a. de rebus naturalibus | ||||||
| i. de rebus aeriis | ||||||
| ii. de terra, solo, agro sim. | ||||||
| iii. de herbis, plantis, arboribus, lignis sim. | ||||||
| 1. de ipsis plantis | ||||||
| 2. de partibus plantarum (fere i.q. offensionibus obnoxius) | ||||||
| a. de fructibus | ||||||
| b. de aliis partibus | ||||||
| iv. de materia corporum | ||||||
| v. de cera, metallis, lapidibus sim. (interdum agitur de rebus, quae e duris tractando molliuntur) | ||||||
| vi. de umoribus tangenti, natanti cedentibus | ||||||
| vii. de harena, pulvere sim. | ||||||
| viii. de materiis accuratius non definitis | ||||||
| b. de rebus ad usum animantium paratis sive natura mollibus sive mollioribus tractando factis | ||||||
| i. de cibis | ||||||
| ii. de vestimentis, vittis sim. | ||||||
| iii. de lectis, feretris sim., quae impositis culcitis al. molliora fiunt | ||||||
| iv. de aliis rebus | ||||||
| c. de animantibus | ||||||
| i. de ipsis corporibus, i.q. ossibus minus duris praeditus, tener | ||||||
| 1. hominum (fere mulierum vel puerorum) | ||||||
| 2. Animalium | ||||||
| ii. de carnosis partibus | ||||||
| iii. de osseis partibus | ||||||
| iv. de lana, capillis sim. | ||||||
| v. de excrementis | ||||||
| 2. i.q. languidus | ||||||
| a. de corporibus animantium (in re veneria i.q. impotens coeundi) | ||||||
| b. de partibus corporum | ||||||
| c. de armis | ||||||
| B. latius, i.q. flexibilis, mobilis | ||||||
| 1. de membris | ||||||
| 2. de motibus corporis | ||||||
| 3. de plantis (opp. lentus) | ||||||
| 4. de rebus arte tractatis (in imag. de verbis tamquam rebus corporeis; audacius de operibus arte fingendis – opp. rigidus) | ||||||
| II. translate pertinet ad ingenia, mores, actiones animantium ac statum rerum |
||||||
| A. de animantibus rebusque ad homines pertinentibus | ||||||
| 1. in bonam partem i.q. mitis, placidus, lenis, obsequens, gratus, tolerabilis, mediocris, blandus | ||||||
| a. de rebus ad homines (deos, bestias) pertinentibus | ||||||
| 1. de actionibus hominum (deorum) | ||||||
| 2. de eventis, condicionibus vitae sim., i.q. iucundus, amoenus, facilis | ||||||
| a. generaliter | ||||||
| b. de eventis iniquis, i.q. tolerabilis, minus durus | ||||||
| 3. de animis, vultibus, affectibus | ||||||
| 4. de dictis, sententiis, mandatis sim. | ||||||
| a. generatim | ||||||
| b. potius i.q. blandus, tener | ||||||
| 5. de carminibus | ||||||
| a. generatim | ||||||
| b. de poesi elegiaca | ||||||
| 6. de genere dicendi vel compositionis, pertinet ad stilum remissum, tenuem, nihil duri asperique habentem; transfertur ad ipsum hominem; in carminibus | ||||||
| 7. de numeris, rhythmis, sonis sim. | ||||||
| a. generatim | ||||||
| b. speciatim pertinet ad harmoniae doctrinam | ||||||
| b. de ipsis hominibus (de umbris mortuorum; de deis; adde de rebus quasi affectibus humanis praeditis, fere i.q. propitius) | ||||||
| c. de animalibus | ||||||
| 1. natura mitibus, pacificis | ||||||
| 2. domitis, mansuetis | ||||||
| 2. in deteriorem vel etiam contemnendam partem, i.q. parum obduratus, effeminatus, dissolutus, iners (fere i.q. lascivus, i.q. nimis misericors) | ||||||
| a. de hominibus | ||||||
| 1. de animis, mentibus, moribus sim. | ||||||
| 2. de actionibus rebusque humanis (animalium) | ||||||
| 3. de ipsis hominibus (dea) | ||||||
| a. generatim; per enallagen; de re pro persona; de civitatibus, quarum incolae effeminatae sunt; de parte corporis, i.q. lascivus | ||||||
| b. speciatim de pathicis, cinaedis sim. | ||||||
| A. adi. | ||||||
| B. subst. | ||||||
| b. de animalibus | ||||||
| 1. i.q. iners, ignavus, timidus, imbellis | ||||||
| 2. i.q. delicatus, offensionum non patiens | ||||||
| B. de statu rerum naturalium | ||||||
| 1. pertinet ad temperiem | ||||||
| a. de tempestate, umbra, caelo; de spatiis temporum; inde de ipsis regionibus miti caelo fruentibus | ||||||
| b. de ventis placidis | ||||||
| 2. pertinet ad vim attenuatam, gradum delenitum | ||||||
| a. de fluviis, undis sim.; sim. de pulsu | ||||||
| b. de regionibus paulum declivibus rebusque leniter curvatis | ||||||
| c. de medicamentis vi leni praeditis | ||||||
| d. de igne leni | ||||||
| e. de variis rebus i.q. modicus | ||||||
| 3. pertinet ad saporem sive odorem | ||||||
FIGURE 2: Structure of OLD2 article mollis
| 1 Yielding to the touch; soft | |
| b soft (with cushions or sim. covering; also with grass) | |
| c (applied to food, usu. as a term of commendation) | |
| d (applied as a comparative term to things normally considered as hard) | |
| e (used to describe the class cephalopoda; cf. Gk μαλάκια) | |
| 2 (of the body or its parts) Lacking firmness, soft, flabby, flaccid |
|
| 3 Soft, tender |
|
| a (as typical of women; also of emasculated males) | |
| b (as typical of youth) | |
| c (of statues, as giving the illusion of real flesh) | |
| 4 Supple, flexible, loose, pliant |
|
| a (of the body or its parts) | |
| b (of other things) | |
| 5 (of persons) Physically weak, feeble |
|
| 6 (of terrain, roads, etc.) Providing easy going |
|
| b (of slopes) gentle (also, of curves) | |
| 7 Easily borne, tolerable, mild |
|
| b (of climate, weather) mild, soft, relaxing (also, of places, regions) | |
| c (of tasks, journeys, etc.) easy to accomplish | |
| 8 Agreeably soft to the senses (other than touch), mild |
|
| b (of speech, verse, rhet. style, etc.) free from harshness, smooth | |
| 9 (of actions) Gentle, soft |
|
| b (of things) gently moving or acting | |
| 10 (of conditions, activities, etc.) Of a calm or peaceful nature, unwarlike |
|
| b (of sleep, etc.) | |
| 11 Gentle, conciliatory, complaisant, kindly |
|
| b (of words, actions, etc.) | |
| 12 Easily influenced, impressionable, susceptible, sensitive |
|
| 13 Effeminate in character, weak, cowardly |
|
| b (of actions, opinions, etc.) lacking firmness, weak | |
| 14 (of animals) Mild in nature, gentle |
|
| 15 Effeminate in appearance or behaviour, womanish; (spec.) denoting a passive partner in male homosexual intercourse |
|
| b (of things or actions) unmanly, womanish | |
| 16 (of movement) Languid, voluptuous |
|
| b (of writings) amorous | |
FIGURE 3
3.1: Two antonyms of mollis
durus: Cic. Off. 1.129.2 ne quid effeminatum aut molle et ne quid durum aut rusticum sit; Prop. 3.1.19-20 mollia, Pegasides, date vestro serta poetae: / non faciet capiti dura corona meo; [Tibull.] 3.4.76 vincuntur molli pectora dura prece
fortis: Dom. Mars. incert. 7 ne foret aut elegis molles qui fleret amores / aut caneret forti regia bella pede; Ov. Met. 3.547-8: ille dedit leto fortes; vos pellite molles / et patrium retinete decus; Lucan 10.133-5 nec non infelix ferro mollita iuventus / atque exsecta virum; stat contra fortior aetas / vix ulla fuscante tamen lanugine malas
3.2 Some philonyms of mollis
Hor. Epod. 11.3-4: amore qui me praeter omnis expetit / mollibus in pueris aut in puellis urere; Mart. 9.59.2 inspexit molles pueros oculisque comedit; Apul. Met. 9.28.8 tam mollis ac tener et admodum puer; Gell. N.A. 19.11.4 v 8 et labra pueri mollia
Plaut. Aul. 422: ita fustibus sum mollior magis quam ullus cinaedus; Catull. 25.1: cinaede Thalle, mollior cuniculi capillo
Catull. 11.5: Arabesve molles; V. Geo. 1.57: molles. . .Sabaei; Mart. spect.1.3: molles. . . Iones
3.3 A xenonym of mollis
Cic. Epist. ad Brutum 17.1.8 cum enim mollius tibi ferre viderer quam deceret virum; Cic. Tusc. 2.41.16 vir natus ad gloriam ullam partem animi tam mollem habebit, quam non meditatione et ratione conroboret? Hor. Epod 1.9-10 an hunc laborem, mente laturi decet / qua ferre non mollis viros; Quint. I.O. 5.9.14 fortasse corpus vulsum, fractum incessum, vestem muliebrem dixerit mollis et parum viri signa
3.4 A philonym of vir
Cato, Agr. pr.4.2 ex agricolis et viri fortissimi et milites strenuissimi gignuntur; Ennius, Annales fr. 7.233 fortibus est fortuna viris data; Plaut. Asin. 558 qui me vir fortior ad sufferundas plagas?; Plaut. Cist. 232 sane ego me nolo fortem perhiberi virum; Ter. Andr. 445 cavit ne umquam infamiae / ea res sibi esset, ut virum fortem decet
FIGURE 4.1
FIGURE 4.2: Texts illustrating senses indicated in Figure 4.1
moll- (1): Cels. De medic. 2.1.13: maximeque in mollioribus corporibus, ideoque praecipue in muliebribus
moll- (2): Plin. N.H. 25.61.1 vetant dari senibus, pueris, item mollis ac feminei corporis animive
moll- (3): Hor. Epod. 14.1: mollis inertia; Ov. Met. 11.648 molli languore
moll- (4): Cic. Off. 1.106: quam sit turpe diffluere luxuria et delicate et molliter vivere; Cic. Fin. 1.37: voluptaria, delicata, mollis . . . disciplina; Sen. Epist. 84.1 [voluptates] molliunt et enervant
moll- (5): Hor. C. 2.9.17-19: desine mollium / tandem querellarum et potius nova / cantemus Augusti tropaea; 4.1.4-7: desine, dulcium / mater saeva Cupidinum, / circa lustra decem flectere mollibus / iam durum imperiis; Prop. 2.1.1-2: quaeritis unde mihi totiens scribantur amores, / unde meus veniat mollis in ora liber
moll- (6): Sen. Contr. 2.1.6: incedentem ut feminis placeat femina mollius; Ov. Ars 1.595-6: si vox est, canta; si mollia bracchia, salta, / et quacumque potes dote placere, place; Juv. 6.63-4: chironomon Ledan molli saltante Bathyllo / Tuccia vesicae non imperat, Apula gannit
moll- (7): Hor. Epod. 11.4: mollibus in pueris aut in puellis urere
moll- (8): Hor. Epod. 12.15-16: Inachiam ter nocte potes, mihi semper ad unum / mollis opus; Petr. Sat. 134.2: mollis, debilis, lassus tamquam caballus in clivo; Lucan 10.133-4: infelix ferro mollita iuventus / atque exsecta virum; Mart. 7.58.5: deseris imbelles thalamos mollemque maritum
moll- (9): Phaed. 4.16.1: tribadas et molles mares (cf. 14 ita nunc libido pravo fruitur gaudio); Juv. 9.38: mollis avarus (cf. 9.40 computat et cevet, 130 numquam pathicus tibi derit amicus)