Maclean The Logic of Physiognomony in the Late Renaissance

Early Science and Medicine 16 (2011) 275-295

www.brill.nl/esm www.b rill.nl/esm

e Logic of Physiognomony Physiognomony in the Late Renaissance Ian Maclean All Souls College, Oxford *

Abstract is article studies the advances made in the logic of Renaissance physiognomy from the state of the subject in antiquity and the Middle Ages. e properties and accidents of the human body are investigated in the context of the signs selected by physiognomers, whether univocal or in syndromes, strong or weak in character, negative or positive, consistent with each other or contradictory. When these signs are translated into propositions, the construction of argument which flows from them is shown to be ut plurimum reasoning, in which an element of quasi-mathematical proto-probability and hermeneutical thinking (in the treatment of ambiguity and obscurity) may be detected. ese allow the question “ is x more likely to be the case than y or z? ” to be answered through a variety of procedures. Renaissance physiognomy is shown to be a discipline in which a novel combination of o f rational procedures come together, and a site of conceptual change in respect of property and accidence. Keywords proprium um, physiognomy, logic, Renaissance, probability, semiology, hermeneutics, propri accidens

Like medical diagnosis and various other divinatory practices, physiognomony is grounded in the interpretation of signs. In an ideal world, the interpretation of such signs would yield indubitable knowledge; but for that to be the case, a number of necessary conditions have to be satisfied in traditional logic. The sign would have to refer to one thing only; it would have to be in a straightforward causal relationship to that thing; it would have always to designate that thing; its absence would have to signify the absence of the thing; it would have to be intelligible * Ian Maclean, All Souls College, Oxford Oxford OX1 4AL, England (ian.maclean@all-souls (ian.maclean@all-souls .ox.ac.uk).

276

I. Maclean / Early Science and Medicine 16 (2011) 275-295

(i.e., it would have to be expressed in a language not of proper but of common terms); it would have to be clear (as opposed to obscure), univalent, complete and immediate. In other words, the sign would have to give access to the truth, the whole truth and nothing but the truth; the truth in this case being a correspo correspondence, ndence, or as the philosophy of the Middle Ages and Renaissance would woul d have it, “an “an adequation of the thing to the intellect.” 1 It will come as no surprise that for reasons which arise not only from the nature of the human subject, but also of the sign itself and even of scholastic logic, this ideal situation does not obtain. The physiognomonic treatises (broadly understood) inherited from the ancient world and the Middle Ages by the Renaissance, while for the most part made up of inventories of signs, refer to a variety of semiological and logical problems, all of which are appropriated and repeated by their later emulators.2 The question therefore therefore arises: was anything written in the late Renaissance which would have surprised the most astute of the ancient or medieval authors, such as pseudo-Aristotl pseudo-Aristotle, e, Anonymus Latinus, Michael Scot, or John Buridan, or which could not have been conceived of by them (even if they did not actually articulate it)?3 I think that there is, and that it is to be found in a drift from logic towards hermeneutics and proto-mathematical probability, as well as in subtle semantic transformations of a number of terms, most notably 1)

All these potential difficulties are mentioned in pseudo-Aristotle, Physiognomonica . On the notion of truth, see Gudrun Schulz, Veritas est adaequatio intellectus et rei : Untersuchungen zur Wahrheitslehre des omas von Aquin und zur Kritik Kants an einem überlieferten Wahrheitsbegri Wahrheitsbegriff ff (Leiden, 1992). 2) Martin Porter, Windows of the Soul: e Art of Physiognomy in European Culture, 1470-1780 (Oxford, (Oxford, 2005) contains an exhaustive bibliography of ancient, medieval and Renaissance texts. 3) Polemon’s Physiognomy from See Simon Swain (ed.), Seeing the Face, Seeing the Soul: Polemon’s Classical Antiquity to Medieval Islam (Oxford, 2007) (with editions and translations of the texts of Polemon, Adamantius, Anonymus Latinus and pseudo-Aristotle); Jole Agrimi, Ingeniosa scientia nature: studi sulla fisiognomica medievale (Florence, 2002); Josef Ziegler Ziegler,, “Text “Text and Context: On the Rise of Physiognomic ought in the Later Middle Ages,” in De Sion exibit lex et verbum domini de Hierusalem: Essays on Medieval Law, Liturgy and Literature in Honour of Amnon Linder , ed. Yitzak Hen (Turnhout, 2001), 159-82; and id., “Philosophers and Physicians on the Scientific Validity of


277

probabilitas , coniectura , certitudo , genus , proprium and accidens ; and the explicitly loose use of terms and logic in the discipline of medicine which were carried over into physiognomony physiognom ony (or borrowed from it, in some cases) which I have discussed at length elsewhere. This article brings together into a single argument the scattered material relating to physiognomony and logic in my previous investigations.4 In a much more extensive study, one might contextualize the changes that occur in this field inside the discourses discourse s and institutions in which they occurred; here I am going to give an internalist inte rnalist account of them, in order to make m ake possible in due course a comparison with the logical analysis employed in physiognomonic texts of later periods. 1. e Problem of Demonstrative Knowledge in Physiognomony Physiognomony and Medicine Physiognomony in the Middle Ages was defined by one of its earliest Physiognomony and most influential exponents, Michael Scot, as a ‘scientia naturae’: that is, a body of knowledge obeying certain rules of argument which is subalternated subalterna ted to natural philosophy philosop hy.. Some commentators referred to physiognomony as a necessary (i.e., demonstrative) science, but John Buridan does not, preferring to associate associa te it with ‘probabilit ‘probability’, y’, by which he means that it has the character chara cter of a ‘scientia a posteriori’, that is, one that relies not on necessary premisses but on ones which are plausible or true for the most part, and an d which depends on effects and not n ot causes.5 When physicians physici ans begin to discuss di scuss physiognomony physiogn omony in the later late r Middle Ages, they stress the fact that the bodily signs accessibl accessiblee to t o the t he senses s enses on which physiognomony depends demonstrate demonstrat e the interaction of soul 4)

Ian Maclean, Logic, Signs and Nature in the Renaissance: e Case of Learned Medi- cine (Cambridge, (Cambridge, 2001), esp. 181-88, 315-19; idem, Le monde et les hommes selon les médecins de la Renaissance (Paris, (Paris, 2006), 95-109. 5) e first quaestio (“utrum (“utrum per physonomiam potest haberi aliquod certum iudicium de moribus hominum h ominum”) ”) in Buridan's quaestiones on on Aristotle’ Aristotle’ss Physiognomy (Oxford, ( Oxford, Bodleian Library, MS Can Misc. 422, fol. 112r) makes this claim: “…quia mores mutantur etiam dicendum est quod de casu sit certum iudicium de moribus actualibus sed tamen de inclinatione naturali ex quo tunc verisimiliter arguitur ad omnes actuales et non consequentia necessaria sed vere similiter probabilis.” I owe this reference to

278


and body, body, and they link this thi s to the theory of temperament temp erament or complexcomple xion, according to which humoral changes affect behaviour and can be perceptiblee to the physician perceptibl physicia n’s gaze and his other senses. se nses.6 This associates physiognomic signs with a causal structure, and dignifies the practice of physiognomony thereby.7 The aspiration to characterize physiognomony as demonstrative led one Renaissance commentator (Jacques Fontaine) to reorganize the text of the pseudo-Aristotelian Physiogno- monica , so as to turn it into an a priori science like the Physics , in which the causes are stated ab initio in the form of principles or premisses. He does however concede that even in this form, physiognomony is a scientia ut plurimum ; that is, it applies only in most cases, and its necessity is therefore not that of a demonstrativ demonstrativee science.8 Like medicine, physiognomony physiognomony comes to be described as an art as well as a science, possessing necessary, true and invariable praecepta , even if it is con6)

Ziegler, “Philosophers and Physicians,” Ziegler, Physicians,” 311 points out that Bartolomeo della Rocca alias Cocles links physiognomony also to celestial causation, which according to him lies behind every natural object in the sublunary world. Pietro d’Abano had produced the same claim two centuries earlier in his Liber compilationis physiognomiae . I owe this point to the anonymous reader of this article for ESM. 7) See Ziegler, Ziegler, “Philosophers “Philosophe rs and physic physicians,” ians,” esp. 310 (quoting (quoti ng Pietro d’Abano’s d’Abano’s definition: “physiognomia est scientia passionum anime naturalium corporisque accidentium habitum vicissim permutantium permutant ium utriusque”; and Jole Agrimi, Ingeniosa scientia nature: studi sulla fisiognomica medievale , Florence, 2002). 8) Jacques Fontaine, Phisiognomia Aristotelis ordine compositorio edita ad facilitatem doctrinae [...] commentariis illustrata brevissimis et propter methodum praespicuam facil- limis (Paris, (Paris, 1611), 5-13: “Aristotelis laborem doctis offerens, methodum resolutivam sub obscuram addiscentibus est sequutus, nos studentes st udentes facilitati dedita opera invenimus eum ordinem, evitantes prolixitatem et diversarum opinionum enumerationem, […] Physiognomia est subalternata Physicae […] satis nobis esse debet si alteram necessitatem quam noster author vocat ex suppositione et ut plurimum posset in hac reperiri”; Guglielmo Gratarolo, Opuscula: de memoria reparanda, augenda, conservan- daque, ac de reminiscentia: tutiora omnimodo remedia, praeceptiones optimae. De prae- dictione morum naturarumque hominum, cum ex inspectione partium corporis, tum aliis modis. De temporum omnimoda mutatione, perpetua et certissima signa et prognostica (Basle, 1554), 75: “physiognomiam scientiam esse necessariam, quo per ipsam praedicimus aptitudines naturales ad affectus vel mores: quo vero per ipsam praedicimus affectus vel mores actuales nec est scientia necessaria, nec firma: verum quia homines h omines plerunque vivunt sensu, et non nisi sapientes vivunt ratione, ideo physiognomia et scientia praedicendi mores actuales et affectus ut in pluribus, quoniam plures p lures appetitu


279

jectural in respect of its practice; physiognomony physi ognomony and its allied divinadivin atory practices come also to be characterized as a combination of ratio (the rules) and experientia (their application to individual cases by a practised exponent).9 2.

Genus ,

proprium, Accident

Renaissance thinkers had new materials at their disposition, including novel medical theories of diagnosis and prognosis, the rhetorically structured dialectic of Rudolph Agricola and Petrus Ramus, and a number of newly available a vailable ancient texts, most notably Galen Galen’’s Quod animi mores and Hippocrates’ Hippocrates’ss Airs, waters, places .10 corporis temperamenta sequantur and They incorporated this richer textual resource in their works, and developed some of its propositions in subtle ways inside the parameters of discussion permitted permitt ed by theological authorities. The interaction of soul and body, and other material determinants which act on the human being such as region, climate, diet, mode of life, race and sex came under renewed scrutiny.11 But such material influences were not seen as wholly determinant, except in the case of animals12: the much-quoted 9)

Girolamo Cardano, De libris propriis , in Opera omnia , ed. Charles Spon (Lyon, 1663), I:144, declares that Physiognomony is a “contemplatio maxime rationalis, iudicium postulans, et participationis dialecticarum regularum Ptolemaei usum exercens”: Ptolemy’s rules are set out in Cardano’s Dialectica ( ( ibid ., ., I:307); also Iodocus Willichius, Physiognomonica Aristotelis latina facta: addita est eiusdem interpretis oratio in laudem Physiognomoniae (Wittenberg, (Wittenberg, 1538), A4r. (reference to “physiognomonica […] experientia et ratione confirmata”); confirmata”); ibid ., ., B2v: “neque physiognomiae praecepta necessaria sunt, sed propensionem quamdam commonstrant.” For a full discussion of relevant parallels with medicine, see Maclean, Logic, Signs and Nature in the Renaissance . 10) On these points see ibid . 11) Fontaine, Physiognomia , 15 gives four possible meanings to “aliud sequi aliud”: “ut “ut effectus suam causam; ut causa secundaria primariam; ordine temporis et proportione; ut causa sequitur instrumentum in agendo,” the last of which applies to soul and body (23): “anima operando sequitur corporis dispositionem, et per consequens ex dispositionibus corporis licet coniicere inclinationes et propensiones animae naturales.” 12) Aphorismo ismorum rum metop metoposcop oscopicor icorum um libe libellus llus unus addaeus Hagecius ab Hajek, Aphor (Prague, 1562), sig. B1v: “Quod ad certitudinem huius artis attinet, id cogitare singuli debent, artem esse qualis Medicina et Astrologia, quae quanquam utraque cultior sit

280


case of the sexually moderate and clever Socrates (whose appearance led Zopyrus the physiognomer to declare that he was by nature lecherous and stupid), showed the exercise of reason could combat c ombat and annul moral propensities. Geographical determination was also not absolute: Scythians might be stupid as a rule, rule , and Athenians clever: but there was at least one example of a Scythian philosopher, philosopher, and at least one asinine Athenian.13 These exceptions within the rule have an effect of loosening the meanings of such terms as ‘genus’ and ‘signum’; it can in extremis reduce physiognomonic interpretation to unique acts of judgement without any general application.14 I shall begin with a brief examination of the first term, before passing to the nature and logic of the second. The zoological parallels used in physiognomo physiognomony ny allow its practitioners to assert that animals are determined in their nature and cannot use reason to liberate themselves from it; in this sense physiognomo physiognomony ny is a necessary science, as it relates to invariable relationships between dispositions and corporeal features. But there is a problem in determining the nature of given features: are they proper to their species (hence part of the formal definition of their substance) or are they accidents, arising from transient and contingent material changes in secondary qualities (accidentia are are both the qualities which are non-essentially accidunt, in eo cum eis communicat. Ingenium, mores, fortuna, et casus quidam non obscure cognosci possunt, et plaeraque signa habemus certissimorum eventuum, et plura profert dies in lucem. Non potest etiam haec ita sibi constans et perpetua esse in hominum naturis, ac in brutorum animantibus, propterea quod haec solo naturae instructu feruntur, illae vero etsi ad quaedam sunt propensae; tamen institutione et educatione plaerumque p laerumque mutantur mutantur..” 13) Fontaine, Physiognomia , pp. 13: “omnes homines ut plurimum sequi solere inclinationes suas naturales”; Cicero, Disputationes Tusculanae , iv.80; Apuleius, Apologia , 24.33; Galen, Quod animi mores corporis temperamenta sequantur , , in Opera omnia , ed. C.G.Kühn (Leipzig, 1821-33), IV:822. e Secretum secretorum attributes the Socrates anecdote to Hippocrates, 14) Camillo Baldi, In Physiognomica Aristotelis commentarii [...] ed. Girolamo Tamborini (Bologna, 1621), 45: “non enim nos de toto genere humano iudicium ferre tentamus,neque de omnibus humani generis passionibus, sed de quibusdam et de certis hominibus tantum […] certo in loco habitantibus, et certas quasdam habentibus conditiones, iisdem cibis vescentibus, et consimiliter corpus exercentibus, ac demum de ipsis individuis similiter, similiter, et non omnes animi passiones iudicamus, sed eas tantum quibus accidit mutare corpus […].” e use of certus as as a synomym of quidam is of


281

associated with substances, and, in more strictly medical terms, symptoms). In zoological treatises, the definition of genus depends on the distinction between proprium (which relates to the essence) and accident (which is only qualitative) qualitative).. The former question may require a species to be defined in defiance of what is commonly accepted (the white crow is a crow, for example; blackness is a quality, and cannot be part of the definition of the species). This may also lead to the designation as an accident of a qualitative feature that none the less seems to be a property (e.g., the fact that t hat man’s man’s hair goes grey); such suc h features are referred referre d to as inseparable accidents. As the sixteenth century progresses, one may detect a greater willingness to accept a merging of proprium and accident. Some theorists even speak on the authority of Galen of ‘propria accidentia’.15 As a result of this development, development , ‘property’ ‘property’ can encompass encompas s certain accidental accident al features in zoological and botanical treatises, and this has consequences not only for the concepts of genus and species, but also for the physiognomonic sign, to which I shall now turn. The locus classicus in in which semiological inference is related directly to physiognomony is set out in Aristotle’s Prior Analytics ii.27 ii.27 (70b 12ff.): [Supposing that] one grants that body and soul change together in all natural affections [...] and also that there is one sign of one affection, and that we can recognise the affection and the sign proper to each class of creatures, we shall able to judge character from physical appearance [...] in the first figure, provided that the middle term is convertible with the first extreme, but is wider in extension than the third term and not convertible with it: e.g., if A stands for courage, B for large extremities and C for lion. en B applies to all of that to which C applies, and also to others, whereas A applies to all that to which B applies, and to no more, but is convertible with B. Otherwise there will not be one sign of one affection.

The condition of univocality is stated, together with that of convertibility (all those with large extremities are brave; all brave animals have 15)

See Ian Maclean, “White Crows, Crows, Graying Hair and Eyelashes: Eyelashes: Problems for NatuNatural Historians in the Reception of Aristotle’s Logic and Biology from Pomponazzi to Bacon,” in Historia: Empiricism and Erudition in Early Modern Europe , ed. Gianna Pomata and Nancy Nancy Siraisi (Cambridge, Mass., 2006), 147-79. Galen refers to “idion symptoma”” (see ibid ., symptoma ., 174).

282


large extremities: this is a propriu proprium m, in logical terms). Several other conditions are implied (clarity; the absence of the sign must betoken the absence of the character trait; the character trait must be the cause of the sign). Under these conditions, the predicate ‘courageous’ is a correct demonstrative inference from the sign ‘having large extremiextremities’.16 Problems Pro blems arise from from this this Aristotelian Aristotelian example. The first of these these is the fact that univocal proper signs of this kind are not in fact found in physiognomony: physiognomo ny: at best, inferences are usually made from a number of signs (indeed, it is said to be folly to rely on one sign alone). 17 The stricter logicians refuse to accept that this is demonstrative in an Aristotelian sense18; but others are willing to consider them so. These writers (among them Sanctorius) accept that the proposition p is a correct 16)

Giambattista della Porta, De humana physiognomonia (Hanau, (Hanau, 1593), 57-59. 17) Pseudo-Aristotle, Physiognomonica , ii, 806b 36ff.: “Generally speaking, it is foolish to put one’s faith in any one of the signs: but when one finds several of the signs in agreement in one individual, one would probably have more justification for believing the inference true.” Fontaine, Physiognomia , 40-45; Baldi, In Physiognomia Aristotelis commentarii , 41; Della Porta, De humana physiognomia , 55 points out that this applies to common signs only, and adds a confirmation of the point from a Galenic source: “uni signorum credere fatuum esse, scilicet communium: sed plura circa unum convenientia pensiculanda esse, et plura testimonia ad unum accommodanda, accommo danda, ut ex iis mox securius iudicium proferatur. Quod est etiam a Galeno confirmatum, qui Physignomonos multum errare crederit, quid uni signorum credant, nisi proprium signum id fuerit.”” See also Galen, De placitis Hippocratis et Platonis , v.5, in Opera , ed. Kühn, IV: fuerit. 257 (the error of o f Posidonius). Cf. Fontaine, Physiognomia , 34: “[…] signa autem sunt aut propria aut communia, quae a Galeno dicuntur propria vel inseparabilia. Siquidem ex communibus signis suas textuerit ratiocinationes aliquis, non erit demonstrativa argumentatio. Nam demonstratio ex propriis et necessariis constat ex Aristotele lib 1 de demonstr. Neque certe rationi consonum est, si cervo cum cane aliquod signum est commune, illud magis significare mores cervi quam canis: ut enim propria signa aliquid peculiare denotant, ita communia commune, quae Physiognomo nullo modo commoda esse possunt.” p ossunt.” 18) Sanctorius Sanctorius, Methodi vitandarum errorum omnium, qui in arte medica contingunt libri quindecim (Geneva, 1630), col. 84 (on the four common signs of pleurisy): «Error logicus committitur, quia quatuor signa pathognomica universaliter sumpta constituunt argumentum a positione consequentis ad positionem antecedentis, veluti si homo, est animal, sed est animal, ergo est homo, quod non valet, est enim in secunda figura ex duabus aff. et multis modis peccat, ut alibi ostendimus, est dolor


283

and demonstrative inference from the syndrome of signs s1,s2, s 3 […]; in the case of courage, for example, we might add to large extremities a loud voice and a hairy chest, and produce a syndrome.19 In medicine, such a syndrome is known as a pathognomonic sign: one which unun equivocally allows the doctor to infer the presence of one illness and no other. The logical consequences of this are spelt out by Cardano, who writes as a s follows: Pathognomonic signs must be either proper or inseparable (i.e., pertaining to an accidental feature of the body which is inalienable), or both […] From a proper sign it is valid to argue as follows: a certain sign is present, so a certain passion (i.e., illness) is present. From an inseparable sign, it is valid to argue in the negative mode as follows: a certain sign is absent, so a certain passion is absent. From signs which are both proper and inseparable, it is valid to argue both affirmatively and negatively […] [in the case of pleuritis] five signs are invoked, all inseparable, none proper: acute fever, stabbing pain, difficulty in breathing, coughing, a hard and dense pulse. All these signs are inseparable, but it is valid to argue in the negative mode in the following way: there is no acute fever, fever, therefore there is no pleuritis; there is no stabbing pain, therefore there is no pleuritis, etc. But none of the signs is proper, so it is not valid to argue as follows: there is an acute fever, fever, so there is pleuritis, because acute fever is a sign of other illnesses. From this it results that there is a proper and inseparable sign, from which it is valid to argue both affirmatively and negatively as follows: all five signs are present, so it is necessarily a case of pleuritis; not all the signs are present, so it is not a case of pleuritis. 20 convertibiles termini, est dolor lateris pungitivus, ergo pleuritis: et caetera limitentur, et coarctentur signa propria, ut infra, si velimus inferre consequentias.” consequentias.” 19) Ibid .: .: “Nota autem, ut aliquid certi dicere possumus duo requiris, unum est, ut plura signa sint, quae idem testentur, testent ur, alterum est, ut nullum quod designat designa t contrarium affectum, vel consequens ad contrarium, adsit, ut dico fortem hunc esse, primum quoniam habet magnas extremitates, latum pectus, pilosumque, pilosu mque, os grande, carnem durum, deinde quoniam in eodem non apparet quid quam eorum signorum, quod vel timorem arguat, vel animi mollitiem, vel avaritiam, quae duo per se quidem non opponuntur fortitudini, at comitari soleat tamen timiditatem.” Different combinations of signs may of course yield a range of different conclusions: [Anonymus Latinus], De diversa hominum natura prout a veteribus philosophis ex corporum speciebus reperta est, cogno- scenda , ed. Antoine du Moulin (Lyon, 1549), passim gives worked examples of this. 20) Cardano, De epilepsia , in Opera , X:398: “de pathognomonicis agendum est quae denotant nobis passiones, ubi suppono signa debere esse, vel propria, vel inseparabilia, vel utrumque: si signa nullum habeant horum trium, frustra adducuntur a medicis; in signo proprio licet arguere affirmative, est illud signum; ergo illa passio; passio ; in signo insepa-

284


There is an inductive implication here: the four signs must be an exex haustive enumeration of relevant signs, as set out in Prior Analytics , ii.23, (68b 29), otherwise the logic does not work. The signs which are described here as inseparable accidents can of course be transient: there are also physiognomonic signs which are of this nature and yet which are effects of moral states (blushing, for example, is a sign of a propensity to shame). Any attempt to accommodate temporality in the logic of signs leads however to considerable difficulties which I shall pass over here.21 The introduction of proper and inseparable (or proprium and accidens inseparabile ) reveals two more problems. The first is that the object of physiognomony (the material human body) is subject to temporal change and is itself radically individual. indi vidual. Its signs are therefore like those of Rhetoric , i.2 (1356b 1ff., likely or probable signs which occur only generally (‘hos epi to polu’: ‘ut multum’, ‘ut plurimum’, ‘magna ex parte’); that is to say, even if generally they are accepted as having a certain referent , they are capable of having a different one. This has two consequences: 1. We may now have to consider the following adaptation of a syllogism in the first figure (i.e., one in which the major and minor propositions are both universal and affirmative):

habeant utrumque, licet arguere, et affrimative, et negative: et quamquam in praesenti morbo [epilepsia] Paulus adducat multa signa quae nihil habeant in se certitudinis, quoniam non sunt neque propria, neque inseparabilia, tamen Galenus adducit tria, quae sunt inseparabilia, nullum tamen est proprium. Sed ex illis tribus simul collectis fit unum signum proprium, et est quomodo in pleuridite; ibi adducuntur quinque signa, omnia sunt inseparabilia, nullum eorum per se est proprium, sed omnes simul efficiunt unum signum proprium: acuta febris dolor pungitivus, difficultas anhelitus, tussis, pulsus durus, et densus. Omnia haec sunt inseparabilia, et ab eis licet arguere negative, non est febris acuta, ergo non est pleuritis; non est dolor pungitivus, ergo non est pleuritis; sed tamen nullum est proprium; neque enim valet, est febris acuta, ergo pleuritis, quoniam febris acuta est etiam in aliis morbis, et sic de aliis. Ex his efficitur unum proprium et inseparabile, a quo q uo licet arguere, et affirmative et negative: sunt omnia haec signa, ergo necessario est pleuritis: non sunt, ergo non est.” See also Maclean, Logic, Signs and Nature , 289-90.


285

A is for the most part B B is for the most part C A is therefore for the most part C This is true logically, logically, as the quantifier (for the most part) is not distinguished from the quantifier (all) when applied to predicates. It is not necessarily true mathematically, mathematically, if we assign finite values to ‘for the most part’ and to the terms A, B and C.22 2. The second problem concerns the rule and exception. If things are only what they are as logical categories for the most part, then the definition by genus and differentiae (e.g., man is a mortal, rational animal) fails; identity ident ity of species become dangerously close to similarity similari ty.. This is clearly perceived by doctors at the time, who refer to ‘for the most part’ definitions (or rather descriptions) descrip tions) to account for the variety within animal species: speci es: or to put it differently, they apply the rhetorical rule of ‘more or less’ in determining normality (being born with six fingers does not make one a monster; but being born with two heads does). The The same rule is applied to zoology by natural philosophers in respect of animal traits such as beaks.23 This threatens the parallels made in physiognomony between animals and humans. 3. Semiological Inferences and Hermeneutics Hermeneutics As well as the object of physiognomony posing problems, there are also problems arising from weak or fallacious semiological inferences. in ferences. These These are enumerated in Aristotle’s various logical treatises. It is legitimate to reason from signs for particular cases as follows: All who have milk mi lk are pregnant pregnan t This woman has milk Therefore she is pregnant.

22) 23)

See ibid ., ., 186-89. Ian Maclean, “Evidence, Logic, the Rule and the Exception in Renaissance Law and

286


But the following syllogism is a fallacy: Pittacus [shows [shows signs of] of ] goodness Pittacus [shows [shows signs of] of ] wisdom Thereforee all wise men are good Therefor And this syllogism, which is more more insidious, is an example of the fallacy of the consequent: (all A is B; therefore all B is A): 24 All who are pregnant preg nant are sallow sal low This woman is sallow Thereforee this woman is pregnant.25 Therefor This last syllogism also demonstrates the problem of plurivalency in signs: sallowness has more than one referent, as does pallor pallor,, for example, which may ma y betoken bet oken a number of moral mora l or physical propensiti propensities: es: melmel ancholy,, fear, love, coldness, ancholy co ldness, anger.26 One might assume that such errors are peremptoril peremptorilyy rejected, but this is not the case. The The logician and medical doctor Claude Aubéry (d. 1596), for example, produces an alternative version of the physiognomonic syllogism, not in relation to bodily features but to the curative properties of plants, which, according to Paracelsus whose disciple he is, have physiognomies. This runs as follows: Coltsfoot is associated with pulmonary congestion Thereforee pulmonary congestion is associated with coltsfoot.27 Therefor This is reminiscent of the weak inferential associations referred to by Giambattista della Porta in his Physiognomonica , and related to a rhetorical rather than logical logi cal argument (or more precisely, enthymeme enthymeme)) of

24)

Aristotle, De sophisticis elenchis , vii, 169b 10f. 25) Aristotle, Prior Analytics , ii, 27, 70a 2ff. 26) Maclean, Logic, Signs and Nature , 148-59, 181-89. 27) Ibid ., ., 131-32: Claude Aubéry, Aubéry, Organon, id est, instrumentum doctrinarum omnium in duas partes divisum, Nempe, in analyticum eruditionis modum, et dialecticam, sive methodum disputandum in utramque partem (Morges, 1584), 151. On Paracelsus and


287

which the standard example e xample comes come s to be the t he following: “Atalanta isn isn’’t a virgin, because she goes wandering in the woods with young men.” Della Porta’s examples are “he is poor, therefore he is obsequious”; or “he is irascible, morose, blunt, therefore also envious.”28 In both cases, the consequence of the first term or set of terms follows necessarily from the theory of character, but may not be supported by physiognomonic signs.29 A final example, which we shall meet again, is the following, taken from Aristotle’s work on sophisms: this man is smartly dressed, or is seen to be wandering around at night: therefore he is an adulterer a dulterer..30 These are examples of faulty inferences: there are also weaknesses in the nature of the sign, arising from the fact that it may not have the required degree of certainty. Doctors and physiognomers concede that their concept of sign, just as their concept of genus, is looser than that of logicians.31 They acknowledge the following semiological problems: 1. Not all the propensities of the human body yield signs, and not all the signs on the human body betoken propensities. Some natural propensities, and all acquired dispositions (such as being a mathematician) produce no signs32; and there are signs which are not caused by propensities, propensiti es, such as insignificant insignifica nt lines on the face and hands. We We may 28)

Della Porta, De humana physiognomonia , 49: “est “est et alter mores coniectandi coniect andi modus, quem nullus ante Aristotelem aggressus est, ut ipse in suis Physiognomonicis testatur: et hic modus syllogisticus est, per quem ex duabus vel pluribus passionibus alias possimus coniectari, et aliae inferuntur, cuius apud Rhetores est usus, Hic pauper est: ergo blandus. Si ex signis hominem cognoscimus iracundum, tristem, et immorigeratum; possumus ex his continuo inferre, hunc etiam invidiae nota carere non posse, sed necessario invidum esse, etsi in eius facie, vel corpore invidiae notae nullae extant, et hunc modum dialectico adscribit, cum ex una conclusione illata, aliam inferat; nam tria illa accipiuntur, ut priora et antecedentia, et quarta q uarta affectio, ut conclusion co nclusion infertur.” infertur.” See also Aristotle, Rhetoric , i.2.8, 1355a-b; Quintilian, Institutio , v.9.12. 29) eophrastus’s Characters do do not have much impact in the Renaiss Renaissance, ance, but various characters as moral types were known through Aristotle’s Aristotle’s Rhetoric , ii.12-14, 1388b ff. 30) De sophisticis elenchis , v, 167b 9f. 31) Emilio Campilongo, Semeiotike , ed. Johann Jessenius a Jessen (Wittenberg, 1601), f.12r; Scipione Chiaramonti, De coniectandis cuiusque moribus et latitantibus animi affectionibus σηµέιωτικη moralis moralis seu de signis (Venice, (Venice, 1625), 2. 32) See Della Porta, De humana physiognomonia , 49. Fontaine, Physiognomia , 12 refers to Matthew 6:16, and the reverse possibility (that hypocrites can counterfeit th e signs of a sad countenance while fasting); Gratarolo, Opuscula , 75 refers to John 7:24 in a

288


link the presence of redundant signs to theories of codes current in the late Renaissance, where they are known as nuls.33 This brings together the hermeneutical practice of deciphering language with that of deciphering bodily signs. A consequence of this is the emergence of two possible approaches to signs, which mark today the difference between phenomenological semiology and that of analytic philosophy: in the one case, all signs are said to be reducible to propositions; in the other, a clear difference is maintained between signs which do not involve a process of mental translation (e.g., measle spots), and those which do (e.g., linguistic utterances). This distinction is often expressed in the Renaissance as one between intelligible and sensible signs.34 Several Renaissance physiognomers physiognomers speak of the body as though as of a text.35 33)

Metoposcopia pia , 1-5, cited in Maclean, Le monde et les hommes , 95-99; idem, Cardano, Metoposco Logic, Signs and Nature , 292n (on nuls). 34) Girolamo Cappivaccius, Opera omnia , ed. Johann-Hartmann Beyer (Frankfurt, 1603), 282: “signum ergo medicum est propositio”; Giovanni Argenterio, Opera (Venice, 1606-7), 101: “porro sive signa, sive indicia, vel notas, vel alio verbo voces, quod rem occultam declarat nihil referre velim. Non enim inter haec nomina eam differentiam agnoscimus, agnoscimus, quam Graeci ponunt inter semeion et tekmerion quod sicilicet, hoc necessarium sit indicium, illud vero non necessarium existat. Atque illud etiam ignorari nolim, latius a nobis signi nomen sumi quam ab Aristot. capiatur. Ille enim tantummodo ab effectis signa sumi docet. Nos vero quicquid potest aliquid eorum quae in corpore nostro fiunt significare, nomine signi donamus. Porro significare est tacite admonere unum ex alio: fit autem tacita haec admonitio ex comparatione rei significandi cum significata, simul enim a[t]que facta est huiusmodi comparatio, deprehendit intellectus quod quaerebat: qua ratione ex fumo significare ignem dicimus. Nam quum notum sit effectus a suis causis nasci, non mirum est si uno cognito aliud protinus animus concipit, id quod est significare, quapropter eadem re utimur ad significandum, indicandum, et demonstrandum, nam ex causa morbi relata ad ea quae facienda sunt in aegrotis elicitur indicatio, ex eadem confecto syllogismo fit demonstratio ad symptomata morborum probanda, quum autem affectionem aliquam congoscere volumus absque expressa ratiocinatione causa huisumodi signum sit. Omnia ergo haec probant ex uno diversa, prout ad diversa referuntur.” On the modern debate, see Maclean, Logic, Signs and Nature , 155. Some commentators class signs by Aristotelian category: see Baldi, In Physiognomica Aristotelis commentarii , 22. Scipione Chiaramonti, De coniectandis cuiusque , 3: “signum est sensibile quippiam, quo existente vel facto, mos certus subest, vel necessario, vel probabili nexu. Dico vero necessario, vel probabili nexu, ut a necessitate abstraham, quae vix unquam in signis morum reperitur.”


289

Individual signs are said in one ancient text to be analogous to letters36; this means that the words they spell out are subject to the same rule as that pertaining pertai ning to language, namely, that communicability communicabi lity is only possible through common signs. If all individual characters expressed themselves through unique signs equivalent to proper names, then there would woul d be no comm communic unicati ation. on. (Thi (Thiss is the third disc discipli ipline-s ne-speci pecific fic meaning of proper we have met). 37 So the signs on the body, although proper or inseparably accidental to the body bearing them, must form part of a common vocabulary vocabulary to yield physiognomo physiognomonic nic information. A subsequent problem arising from common signs is the fact that they do not form part of a demonstration, for which convertible proper signs are needed, as has already been said. So there is a potential confusion of grammar, logic and hermeneutics on this point. 38 2. A second problem, also analogous to linguistic analysis, arises from ambiguity and obscurity obscurity.. Ambiguous meaning in utterances arises either from plurivalence in words or from different construals constr uals of grammatical relations; the same sa me can apply to the signs on the body bod y. Individual signs may betoken more than one thing, or in combination they may yield several senses. Obscurity is also a common problem. In both cases, protocols of disambiguation or clarification have to be established. establi shed. This would be easier in the case of bodily signs, if they all had a simple causal relationship to that of which they are a sign (large extremities cause bravery); but they may also al so arise from an effect; or from both cause and effect.39 36)

[Anonymus Latinus], De diversa hominum natura , 3: the parallel with letters and atoms derived from Lucretius is found in medieval texts discussing atomism, and it may well have theologically dangerous undertones. Zenon Kaluza, “Le De universali reali de Jean de Maisonneuve et les Epicurei litterales,” Freiburger Zeitschrift für Phi- losophie et eologie , 35 (1986), 465-516; also idem, Les querelles doctrinales à Paris: nominalistes et réalistes aux confins du XIVe et du XVe siècles (Bergamo, 1988); Luca int ellectuelle à l’Université de Paris (XIIIe-XIVe (XIIIe-XIVe siècles) (Paris, Bianchi, Censure et liberté intellectuelle 1999). Letters (from the Psalms) are also used in the interpretation of dreams (see A ll Souls College, MS 81, f. 211-2). 37) Physiognomica , i, 805b 22: “if anyone were to pick out the individual characteristics characteristics of each animal, he would not be able to explain of what these are the signs”; Fontaine , Physiognomia , 34, cited above, note 16. 38) Fontaine, Physiognomia , 35-6 39) Chiaramonti, De coniectandis cuiusque moribus , 3 divides signs as follows: “causa

290


3. Except in the case of proper, (convertible) signs, the absence of a sign does not necessarily necessaril y betoken the absence of a propensity propensi ty.. A passage from Aristotle’ A ristotle’ss Rhetoric to Alexander , xii (1430b 30ff.) makes some of these points: One thing is a sign of another—not any casual thing of any other casual thing, nor everything whatever of everything whatever, but only a thing that normally precedes or accompanies or follows a thing. Something happening may be a sign not only of something happening but also of o f something not happening, and something that has not happened may be a sign not only that something is not a fact but also that something is a fact. A sign may produce either opinion or full knowledge; the best kind of sign is one that produces knowledge, but one that causes an extremely probable opinion is the second best kind.

Here mention is made not only the possibility of signs bearing beari ng negative information, and of absence or omission being possibly a sign of something positive, but also of a hierarchy of certainty, to which I shall return. 4. We shall have also to come to grips with the possibility of the confluence of contrary signs in a given case. A number of principles can be adduced to deal with such contrary indications. The first is to suspend judgement, a course recommended by at least one text.40 The second is to believe the greater number of mutually consistent signs. The third is to give credence to the stronger signs (‘signa potentiora’). Various hierarchies are provided to enable this choice to be made, the most common being that the preference is to be given to signs of and close to the eyes; thereafter the face; and in last place those associated with the abdomen. abd omen.41 A final category of sign is the sign or selection of

latentis]; effectus effect us infert effectum [signum rerum consensus consens us [sive] sympathia].” Cf. Caspar Peucer, Commentarius de praecipuis divinationum generibus (Wittenberg, 1553), f. 60r: “[signa] sive causae significatorum, sive effectus, sive causis effectibusve copulata.” copulata.” 40) Pseudo-Aristotle, Physiognomonica , ii, 806a 25ff.; Camillo Baldi, In Physiognomica Aristotelis commentarii , 44: if signs are not “certa et habent contrarietatem aliquam inter se, tali casu suspendendum est iudicium, nisi alterum alter um signum altero praevaleat.” 41) Ibid ., ., 46, (from eyes to abdomen); another hierarchy of signs is given on p. 41: “signa quae sumuntur ab apparentia, quam vocant morem, et a motibus, et a figura corporis, omnino aliis sunt efficaciora, vel quoniam minus ad voluntate pendent, vel


291

signs designated by an authority (a seasoned physiognomer) to be dominant in any one act of judgement.42 This has a medical parallel in the practice of indication. indi cation. The doctor who has to decide the treatment, the strength of the dosage and the frequency of its administration does so by an act of intuition informed by his knowledge of the precepts of the medical art and his practical experience.43 The physiognomer deals with equivocal information information in the same way; or he follows the conditional logic set out in a worked example by Anonymus Latinus, who uses one sign to attenuate the force of another: is man has the eyelids of a chatterer, but the brow of a thinker and the eyes of a madman. e brow does not allow him to be a complete chatterer and the eyelids detract from the thoughtful brow; then again the fierceness of the eyes rather affects the state of the brow brow.. According to these things, the man is judged less of an inopportune chatterer and impetuous character rather than obviously mad. 44

4. Proto-probability We have now reached a point where logic, which has already in one We case given way to hermeneutics, now gives way to an inchoate version of probability: in other words, the question is not “what is entailed, implied or presupposed by x” but “is x more likely y or z?” 45 Such cuntur, faciei enim appraentiam pro libito non mutabis, neque corporis formam, aut saltem maxima cum difficultate. difficultate.”” 42) Fontaine, Physiognomia , 46. 43) Maclean, Logic, Signs and Nature , 306-14. 44) [Anonymus Latinus], De diversa hominum natura , 11. 45) Camillo Baldi, In Physiognomica Aristotelis commentarii , ch. 5, 41-46: “qua ratione procedendi sit in iudicando et quibus signis maior fides, quibus minor sit adhibenda?” In a complex and dense exegesis of the pseudo-Aristotle, Physiognomonica , ii, 806b 36ff., Baldi says that it is impossible for there to be opposite affections in the same subject at the same time, or for signs to be perfectly equal in force. He asserts that more than one sign must be invoked, sets out two orders of signs (‘certiora’ and debiliora’), debiliora’), and a number of rules, among which are the following: “1. cui inest apparentia consequens actum talis passionis, ille inest propensio ad talem actum et operationem secundum illam passionem et dispositionem; 2. Quorum hominum corpora, vel actiones aliquod certum animal referent, illorum etiam animi propensiones et mores eiusdem animalis imitantur; 3. Cum duo apparebunt signa contraria, quorum unum altero sit

292


probability exists in a number of forms in relevant medical and physiognomonic texts: text s: relative frequency frequen cy,, dominance, some som e presence of error, inexactness, approximation, approximation, arguments from suspicion and presumption; in other words, forms of deductive or non-deductive inference from complex evidence which take into account its incompleteness incompleteness.. I have not got the space to review all of these possibilities, but wish here to draw attention to the act of bricolage by which a judgement is reached about the strength of the evidence for a given conclusion. In medicine, the anti-Galenist Giovanni Argenterio (1513-1572) provides an example of a quasi-quanti quasi-quantified fied rule for judging the relative strength of prognostic signs against their number: If an equal number of signs are found, those which have the greater force determine the prediction; if the force of the signs is equal, then the greater number offers the certain indication of what will happen; but if an equal number of good and bad signs are mixed together, then recourse will have to be had, as Galen says, to an experienced medical practitioner, who has frequently engaged in prediction and will somehow recognize the hidden powers of the eventual victor and what prevails in given diseases. 46

The patient will not languish because the experienced doctor, unlike Buridan’s ass, will always be intuitively inclined to one interpretation of the evidence over all others. 4. Si duo signa aequalia, et aequaliter apparentia, et oppositas passiones designantia in eodem subiecto reperientur aliquando, quorum unum sit prioris ordinis, et alterum posterioris secundum indicationem signi, quod est prioris ordinis iudicabimus.” [e ordo is given on p. 46: “[plura signa posterioris ordinis] certa, stabilia et simul convenientia, superant illud, quod quamvis q uamvis sit primi ordinis, tamen solitarium est.” But But it is also true that “plura paucioribus semper, semper, apparentia magis iis quae q uae minus apparent et certiora act pollentiora incertioribus, et debilioribus praecellunt […].” [Anonymus Latinus], De diversa hominum natura , 10 has a rather different list of rules (“multa paucis praeponenda sunt; clara obscuris; potiora minoribus”). 46) Argenterio, Opera , col. 1780: “quod si par numerum [signorum] reperiatur, quae maiorem vim habent, praedictionem attrahunt: quod si aequalis signorum vis fuerit, maior numerus certa praebet futuri indicia, at si bona malis permixta vi et numero paria videantur, videantur, exercitato in artis operibus viro opus est, inquit Galenus, qui ea saepenumero sit contemplatus, latentesque adhuc quodam modo vicentis vires, et quid in singulis morbis praevaleat, praenoverit.” Camillo Baldi, In Physiognomica Aristotelis commentarii , 46 reports Agostino Nifo as saying that it is impossible for two signs


293

The text which first expressly submits all these logical problems to a primitive version of quantified analysis and transforms the standard examples is a book on physiognomony published in 1625 by Scipione Chiaramonte (1565-1652), who taught mathematics at Perugia and, later, philosophy at Pisa. There, the author gives the much-quoted example of the sign of pallor betokening a finite number of things (melancholyy, fear, cold, anger, amorous passion), and shows how a pal(melanchol lid individual can be interpreted as being increasingly likely to be suffering from one of these states by a process of elimination; if four of the five could be eliminated this would allow one to be certain as to how to interpret the sign. This logical process can also be transferred to cases of infinite probability, prob ability, where the common signs of a condition conditio n can form a syndrome of signs which is persuasive. If (in the standard example I have quoted) a woman has breast milk, it may be the case that she is pregnant; if she also has amenorrhea, bizarre appetites, frequent vomiting, and is lethargic, she is more likely to be pregnant. Chiaramonte points out that most physiognomonical signs sig ns are common or equivocal, and have to be conjoined with others to yield even probable information (e.g., wandering around around at night is a sign of an adulterer, a thief, a sleepwalker, and no doubt other things; wearing fine clothes clothe s is a sign of an adulterer adulte rer,, a rich man, an actor, and no doubt other things; but wandering around at night in fine clothes is more likely a sign of an adulterer than of anything else). 47 This example is to be a 47)

Chiaramonti, De coniectandis cuiusque moribus , 5-6. e sequence of reasoning runs as follows: the fully satisfactory case is of a sign as cause, in the first figure: whoever exposes his life for a friend truly loves him; Pylades exposed his life for Orestes; therefore Pylades truly loves Orestes. In the second figure, the sign cannot be used convertibly: whoever is afraid is pale; John is afraid; therefore he is pale; and it can even give rise to paralogisms: Hector is strong; Hector stutters; therefore all stutterers are strong. But the paralogism of the last example can be eliminated in the case of a perfect induction (i.e., one which covers every individual case, showing that Hector, Achilles and Samson, the only three men who can be called strong, all stutter); and there are ways of increasing the probability of the second case. Let us say that pallor is a sign of a person who is afraid, a person in love, a person who is cold, a person who is angry, and a person who is falling fallin g sick (and nothing else). else ). e syllogism: pallor pallo r is a sign of fear; John is pallid; therefore John is afraid ‘lacks ‘ lacks necessity’ and only has ‘refracted probability’,’, according to Chiaramonte; but if you can show that John is not in love, is probability

294


adapted by Chiaramonte from the Aristotelian sophism I have already mentioned above (p. 287), and shows the proximity of this sort of reasoning to modes which were universally universal ly seen to be erroneous at this time. The same form of reasoning reasoning can be found in diagnostic diagn ostic writing.48 5. Conclusion I am now in a position to offer an answer to the question: what would Aristotle and Buridan have found surprising about the discussions of the logic of physiognomony physiognomon y at the end of the Renaissance? Renaissan ce? They would certainly have been aware of the ut plurimum character of the art, and the problems arising from the hierarchy of signs; what might have

(and growing) degree of probability is reached” (“quod, (“quod, si ostendatur, qui pallet, non amare, gradus aliquis accedit probabilitatis. Crescit autem, si neque aegrotare monstretur. Quod, si alia omnia membra; praeterquam timoris tollerentur, argumentum evaderet ex numeratione sufficienti partium. […] vagari noctu est signum commune adulterii, sed convenit et aliis, ut furi. Ornatus vestium est signum et ipsum adulterii commune. Singula debilem fiem adulterii faciunt, at simul congesta maiorem, quo plura vero coniungimus signa, eo magis intenditur inten ditur vis rationis.” An interesting medical parallel is offered by Giambattista Giambattist a da Monte, Medicina universa , ed. Martin Weindrich (Frankfurt, 1587), i.107: “argumentum a signis consequentibus verisimilibus, et ubi multa signa concurrunt, suspicionem augent, licet certitudinem non faciant. Ita in hoc nobili iuvene [the subject of the consultation about a patient possibly suffering from syphilis] est suspicio accessisse aliquam infectionem primo ex bubonibus, et ulceribus pudendorum quae in omnibus solent contingere ante morbum gallicum licet non vertatur; ita ut non in omnibus qui bubones et ulcera pudendorum patiuntur, morbus gallicus consequatur. Accedit gravitas capitis prius non percepta, quae sequitur ad eum morbum et curavi multos, et credo Clariss. hos Doct. multos etiam vidisse qui nullum aliud symptoma habebant praeter dolorem capitis, et tamen laborabant morbo gallico. Praeterea in ulceribus est adustio et nigredo, quod si furunculi quos patitur, sedem etiam duram seu basim solidiorem haberent, augeretur magis suspicio, ita ut devenirem ad certiorem coniecturam. Hinc habeo magnam suspicionem ex o mnibus simul additis de morbo gallico [...].” 48) Sanctorius Sanctor ius Sanctor Sanctorius’ ius’ss Methodi vitandarum errorum , cols. 57-58, is not presented as a mathematical approach, but by employing six sources of signs and using them as a means of eliminating possibilities, it operates o perates very much as Chiaramonte’ Ch iaramonte’ss rising scale of certainty. Sanctorius claims here that it will provide indubitable results, but elsewhere (cols. 94-116) only that it will provide the most probable result. See Maclean,


295

surprised them were the consequences for zoology (the rethinking of accidence and property), the application of hermeneutics herme neutics and rhetorical rhetorical analysis (not in the service of persuasion, but of argument and approximative truth), and the new proto-quantified version of probability. The symptomatic word groups which reveal these changes are those concerning genus, accident and proprium; those concerning concerni ng the exception and the rule; and those concerning concerni ng certainty, conjecture and probability.. What is specially striking, is the competing senses of given words ability in different disciplinary settings (proprium for example, or probabilis or certus or coniectura 49); as physiognomony is a place where several disciplines meet, it becomes a privileged site of conceptual change in the late Renaissanc Renaissance. e.50

49)

Fontaine, Physiognomia , passim and 42: “signa magis perspicua et certiora”; 45: “collectionem [signorum] rationi consonam et probabilem”; Camillo Baldi, In Physiog- nomica Aristotelis commentarii , 41-47 has many uses of certus and probabilis showing semantic variation or slippage; also Della Porta, De humana physiognomia , 55 (‘fixis signis’). [Anonymus Latinus], De diversa hominum natura , 3 (du Moulin’s Moulin’s reference referenc e to ‘verisimila coniecturae’) and Johannes de Indagine, Fisionomia con grandissima brevità da i libri di antichi filosofi , ed. Antoine du Moulin, trans. Paolo Pinzio (Lyon, 1550), ff. 3-4 on the uncertainty of the t he sciences of his day, day, but the need to make conjectures (without which America would not have been discovered). He associates the revival of physiognomony and chiromancy to this willingness to make guesses. On the change is some of these word groups, see Ian Maclean, “Expressing Nature’ Nature’s Regularities and their Determinations in the t he Late Renaissance,” in Natural Law , ed. Lorraine Daston and Michael Stolleis (Aldershot, 2008), 29-44. On ‘gradus certitudinis’, see Maclean, Logic, Signs and Nature , 183n. On conjecture, see ibid ., ., 290-91. 50) I have not addressed the questions about the change from the Middle Ages to the Renaissance suggested by Ziegler, “Philosophers and Physicians,” 312 (did the relegation of the Secretum secretorum to the sidelines of scholarly discourse play a significant role in undermining physiognomony’s status as a valid science? Where was physiog-

Copyright of Early Science & Medicine is the property of Brill Academic Publishers and its content con tent may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Maclean The Logic of Physiognomony in the Late Renaissance

Recommend Documents