And the countenance of an angry or sorrowful man indicates the feeling in his mind, independently of his will: and in the same way every other emotion of the mind is betrayed by the tell-tale countenance, even though we do nothing with the intention of making it known. This class of signs, however, it is no part of my design to discuss at present. But as it comes under this division of the subject, I could not altogether pass it over. It will be enough to have noticed it thus far. Conventional signs, on the other hand, are those which living beings mutually exchange for the purpose of showing, as well as they can, the feelings of their minds, or their perceptions, or their thoughts. Nor is there any reason for giving a sign except the desire of drawing forth and conveying into another’s mind what the giver of the sign has in his own mind. We wish, then, to consider and discuss this class of signs so far as men are concerned with it, because even the signs which have been given us of God, and which are contained in the Holy Scriptures, were made known to us through men—those, namely, who wrote the Scriptures. The beasts, too, have certain signs among themselves by which they make known the desires in their mind.

For when the poultry-cock has discovered food, he signals with his voice for the hen to run to him, and the dove by cooing calls his mate, or is called by her in turn; and many signs of the same kind are matters of common observation. Now whether these signs, like the expression or the cry of a man in grief, follow the movement of the mind instinctively and apart from any purpose, or whether they are really used with the purpose of signification, is another question, and does not pertain to the matter in hand. And this part of the subject I exclude from the scope of this work as not necessary to my present object” (Augustine, On Christian Doctrine, II, 1–3, online trans. by J. F. Shaw, from Select Nicene and Post-Nicene Fathers, at http://www.ccel.org/ccel/augustine/doctrine.iv.iii.ii.html). On the doctrine of the sign in Augustine, see Manetti (1987), Vecchio (1994), and Sirridge (1997).

23. “Liquet autem ex suprapositis significativarum vocum alias naturaliter, alias ad placitum significare. Quecumque enim habiles sunt ad significandum vel ex natura vel ex impositione significative dicuntur. Naturales quidem voces, quas non humana inventio imposuit sed sola natura [contulit], naturaliter [et non] ex impositione significativas dicimus, ut ea quam latrando canis emittit, ex qua ipsius iram concepimus. Omnium enim hominum discretio ex latratu canis eius iram intelligit, quem ex commotione ire certum est procedere in his omnibus que latrant. Sed huiusmodi voces que nec locutiones componunt, quippe nec ab hominibus proferuntur, ab omni logica sunt aliene” (Petrus Abaelardus, Dialectica, First complete edition of the Parisian manuscript, With an introduction by L. M. De Rijk, Ph.D., Second, revised edition, Assen: Van Gorcum 1970, p. 114).
24. “Significare Aristotelis accipit per se intellectum constituere, significativum autem dicitur, quidquid habile est per se ad significandum ex institutione aliqua sive ab homine facta sive a natura. Nam latratus natura artifex, id est Deus, ea intentione cani contulit, ut iram eius repraesentaret; et voluntas hominum nomina et verba ad significandum instituit nec non etiam res quasdam, ut circulum vini vel signa quibus monachi utuntur. Non enim significare vocum tantum est, verum etiam rerum. Unde scriptum est: nutu signisque loquentur (Ovid, II Trist. 453). Per ‘significativum’ separat a nomine voces non significativas, quae scilicet neque ab homine neque a natura institutae sunt ad significandum. Nam licet unaquaeque vox certificare possit suum prolatorem animal esse, sicut latratus canis ipse esse iratum, non tamen omnes ad hoc institutae sunt ostendendum, sicut latratus est ad significationem irae institutus. Similiter unaquaeque vox, cum se per auditum praesentans se subgerat intellectui, non ideo significativa dicenda est, quia per nullam institutionem hoc habet, sicut nec aliquis homo se praesentans nobis dum per hoc quod sensui subjacet, de se dat intellectum, sui significativus dicitur, quia licet ita sit a natura creatus, ut hoc facere possit, non est ideo creatus, ut hoc faciat. ‘Significativum vero magis ad causam quam ad actum significandi pertinet, ut sicut non omnia significativa actualiter significant, ita non omnia actu significantia sint significativa, sed ea sola quae ad significandum sunt instituta” (Logica “Ingredientibus,” Glosses on the Peri Hermeneias. In Bernhard Geyer, Peter Abaelards Philosophische Schriften, Münster, Aschendorff, 1927, pp. 335–336).
25. An explanation of why in the case of animals nature acts as a sort of agent will (comparable to “agent intellect”) is provided by Albertus Magnus, De anima II [De voce qualiter fiat], iii, 22: “Et cum duo sint in anima, affectus scilicet doloris vel gaudii et conceptus cordis de rebus, non est vox significans affectum, sed potius conceptum. Cetera autem animalia affectus habentia sonos suos affectus indicantes emittunt et ideo non vocant; et quaecumque illorum plurium sunt affectuum sunt etiam plurium sonorum, et quae levioris sunt complexionis, et ideo aves plurium sunt garrituum quam gressibilia. Et illae quae inter aves sunt latioris linguae, et melioris memoriae, magis imitantur locutionem et ceteros sonos, quos audiunt. Licet enim bruta habeant imaginationem, sicut superius ostendimus, tamen non moveretur ab ipsis imaginatis secundum rationem imaginatorum, sed a natura et ideo omnia similiter operantur; una enim hirundo facit nidum sicut alia, et haec imitatio est naturae potius quam artis.Ideo anima imaginativa in eis non regit naturam, neque agit eam ad opera, secundum diversa imaginata, sicut facit homo, sed potius regitur a natura et agitur ad opera ab ipsa, ei ideo fit quod licet habeant apus se imaginata, tamen ad exprimendum illa non formant voces. Affectus autem laetitiarum et tristiarum magis profundatur in natura quam in anima, et ideo illos exprimunt sonis et garritibus.”
26. According to this reading the affections of the soul are not mental images of things, but modes of being of thought, cognitive modalities (like thinking, being afraid, or experiencing joy). The pragmata are not real things or facts in general (otherwise how could we explain why Aristotle says elsewhere in his works that we can think of nonexistent or false hybrids like the hircocervus or phenomena whose existence we are unable to prove (such as squaring the circle or the commensurability of the diagonal). Likewise, “those things that are in the voice” could be transformations, differentiations, or articulations that are proper to the human voice. Finally, an expression like kata syntheken does not mean that linguistic voces are related to the affections of the soul by means of convention, but that they are articulate, the effect of a syntactic composition (and for this very reason the voces emitted by animals, which are inarticulate, cannot express thoughts). At the same time the interpretation of the omoiomata is also called into question. It would refer to the fact that there exists a relationship of structural similarity between logical-cognitive operations and events in the world. In conclusion, the passage from Aristotle should be reinterpreted as follows: “The articulations of the human voice and the logical-cognitive operations of the human soul are different from each other and complementary, just as written articulations and articulations of the voice are. And just as the minimal units with which and in which writing is articulated are not the same for all mankind, neither are the minimal units in which the linguistic voice is articulated. On the other hand, the logical-cognitive operations of which the vocal and graphic units are the natural physiognomic signs are the same for all mankind, and likewise the same for all mankind are the facts with which the logical-cognitive operations of the human soul are in a relation of similarity” (Lo Piparo 2003: 187).
27. De Interpretatione, in Aristoteles Latinus II, 1–2, ed. L. Minio-Paluello, Bruges-Paris, Desclée de Brouwer, 1965, pp. 5, 4–11 and pp. 6, 4, 11–13. The following is a translation of Aristotle’s original Greek text: “Now spoken sounds are symbols of affections in the soul, and written marks symbols of spoken sounds. And just as written marks are not the same for all men, neither are spoken sounds. But what these are in the first place signs of—affections of the soul—are the same for all; and what these affections are likenesses of—actual things—are also the same … A name is a spoken sound significant by convention, without time, none of whose parts is significant in separation … I say ‘by convention’ because no name is a name naturally but only when it has become a symbol. Even inarticulate noises (of beasts, for instance) do indeed reveal something, yet none of them is a name.” Aristotle’s Categories and De Interpretatione, Translated with Notes and Glossary by J. L. Ackrill, Oxford: Clarendon Press, 1963, pp. 43–44.
28. Boethius translates semeion as nota on his own initiative, whereas he finds the identification of symbolon with nota already sanctioned by Cicero (Topica VIII, 35), on whom he comments as follows: “Nota vero est quae rem quamque designat. Quo fit ut omne nomen nota sit, idcirco quod notam facit rem de qua praedicatur, id Aristoteles symbolon nominavit” (In Topicis Ciceronis Commentaria IV, PL. 64, col. 1111 B). And here Boethius establishes the equivalency, as characteristic properties of nota, between rem designare and rem notam facere, in other words, between the significative function proper to Aristotle’s symbolon and the inferential or symptomatic function of semeion.
29. See Latratus canis (“On Animal Language”), p. 29, n. 20, and De Resp. 476 a 1–b 12; Hist. An. 535 b 14–24; De an. 420 b 9–14. And along the same lines Boethius, In Librum Aristotelis De Interpretatione Commentaria majora, PL 64, col.