Is 'every man an animal true' when no man exists?

'EVERY MAN IS AN ANIMAL' - EARLY WRITING

Main (see also The Square of Opposition).

Introduction
Richard the Sophist – every phoenix exists.
Robert Kilwardby - whether 'a man is an animal' is true, even though no man exists.
Anon - whether every man of necessity is an animal.
William of Sherwood – passages from Introduction to Logic, and Syncategoremata.
Roger Bacon - from the the Summulae Dialectices.

Introduction

The passages here represent early views (from approximately the 1230's to the 1260's) on the question of whether the A proposition 'a man is an animal' is true when no man exists. Authors include Sherwood and Kilwardby (who thought it is true when no man exists), and Bacon (who thought it isn't).

By implication, those who held that the proposition is true when the subject term is empty, would have held that the corresponding O proposition (some man are not animals) is false. This presents an obvious difficulty for the Moody-Klima-Parsons thesis that, for most of the history of Aristotelian logic, logicians assumed that negative particular propositions are vacuously true if their subjects are empty.

Source material has been difficult to find – in one case I have had to supplement parallel Latin-English with an English translation of Kretzmann, in another I could only locate a brief quotation from Bacon, probably the most important writer represented here. As the basic principle of the Logic Museum is no translation without the translated text itself, I will try to locate the original sources where appropriate, and translate into English, where necessary.

Richard the Sophister

Richard the Sophister (Richardus Sophista) was an English philosopher/logician who studied at Oxford, probably during the second quarter of the thirteenth century. Richard's identity is uncertain, but he is known to be the author of a collection of logically puzzling sentences, sometimes called “sophisms”, entitled Abstractiones. The Stanford Encyclopedia of Philosophy article has more here.

The passage here was probably written in the 1230's or 40's. Richard argues that 'every man of necessity is an animal' has at least two senses, depending on the scope of 'of necessity'. When it applies to 'every man', it says that every man is such that, by necessity, he is an animal, i.e. Plato is necessarily a man, Socrates is necessarily a man, &c. In this sense it is false, because Socrates is not necessarily a man (he may possibly cease to exist). When it applies to the whole sentence, i.e. when it says that the sentence 'every man is an animal' is necessarily true, the sentence is true.

I have translated the Latin term esse habitudinis as 'conditional being'. This is the being of the subject when a universal proposition (every man is an animal) is read as a conditional, i.e. 'if [anything] is a man, it is an animal'. Kretzmann (in his translation of William of Sherwood below) renders it as 'relational being'.

It is thought by some that Richard was actually Richard Rufus. See The Richard Rufus of Cornwall project.

Robert Kilwardby

Robert Kilwardby (c 1215 – 1279) composed a number of works on logic and theology. He was a student in Paris in the 1230's. He is mainly known for his role in the Oxford condemnations of March 1277, following Tempier's condemnation in Paris. Kilwardby prohibited thirty propositions, fourteen concerning grammar and logic, the rest concerning natural philosophy. According to Broadie, among these were those claiming that a necessary truth requires the perpetual existence of its subject (allegedly false, because God's knowledge of contingent truths is itself necessarily true), and that only what exists can be the subject of a demonstration (allegedly false, because it is possible to formulate a demonstration about a non-existent essence).

The quotations here, probably written around 1250, are taken from a secondary source (Henry loc. cit.), who took them from Ebbesen. They are not definitely attributable to Kilwardby, but are certainly consistent with his position. The author argues that 'a man is an animal' is true even when no man exists, because, just as in reality someone being a man is related to his being an animal, in the same way the thought (intellectus) of his being a man is related to the thought of his being an animal. Therefore 'a man is an animal' is true in virtue of the thought, rather than the reality, and therefore it is true even though no man exists. (Just as we now say that a proposition is analytically true, or that it represents a necesary relation between concepts, though Kilwardby would not have expressed himself in this way).

This was a reasoning used by later philosophers to explain why 'essential propositions' must be eternally and necessarily true. A counter-argument, given by Boethius of Dacia [link to follow], was that 'composition in the understanding' (such as the composition of the concepts man and animal in the thought man is an animal) is only true because there is composition in reality. The thought that a man is an animal, is only true if some real man is an animal. Thus 'a man is an animal' is false when no man exists, because there is no corresponding composition in reality. [link to follow].

Anon

The anonymous quotations are from a manuscript Omnis homo de necessitate est animal, ms. Paris, Nat. lat. 16135, fol. 11rb-12vb, written c. 1250. My source was De Libera 1991. The author mentions the fact that there is disagreement about whether the proposition is true when no men exist, and also mentions the distinction between essential being (as in 'a dragon is fire breathing') and actual being (a fox is in the garden). It also mentions the idea of esse habituale - conditional or relational being. 'Every dragon breathes flame' is true in the sense that it predicates conditional being – if any dragon exists, then it breathes flame, which is true whether or not a dragon exists. A corollary of this view is the distinction between the use of 'is' as an absolute predicate (e.g. 'Caesar is', i.e. Caesar exists), and its use as a mere copula, as in 'Caesar is a man'.

William of Sherwood

William of Sherwood (c. 1200-66)was a master at Oxford in 1252, and is known mostly for two works on logic, the Introduction to Logic (c. 1250) and the Syncategoremata (c. 1250). The two passages here are taken from these. Sherwood argues that 'every man is an animal' has two senses, sometimes meaning things that actually exist, in which case it is false when no man exists, sometimes meaning that a certain nature (being an animal) is suited naturally to be in a certain kind of thing (man), even though it is not actually in one. In the latter sense, it has the force of the conditional 'if a man is [i.e. exists], an animal is', and thus is true even when no man exists.

He is quite explicit that a term can 'supposit for' (roughly: denote)something that does not exist – e.g., if there are only two men 'every man exists' is false (based on the bizarre idea that if there are only two men 'every man exists' is false, in which case its contradictory, 'some man does not exist' is true). Therefore a man does not exist. But nonexistence is not predicated of something that exists; therefore 'man' supposits for something that does not exist.

It seems to be the view roundly criticised in Ockham's Summa Logicae II.4 (also to appear in the Logic Museum – link to follow), although Ockham's attack may also be aimed at Richard the Sophister.

Roger Bacon Roger Bacon (c 1214 – post 1292) is an important figure of the thirteenth century. He was at Oxford in the 1230's, at Paris in the 1240's, later returning to Oxford where he remained until the 1290's. He is thought to have anticipated and possibly influenced the nominalistic positions taken up later by Scotus and Ockham [link to follow].

The passage here is quoted in Henry (p. 265) who has taken it from a secondary source (Ebbesen & Pinborg – see bibliography). It is probably from the the Summulae Dialectices, which De Libera thinks was written at Oxford around the 1250's. This is a mature work, which contains two important doctrines: the doctrine of univocal appellation, and the doctrine of predication in respect of empty terms. Bacon argues that a word cannot apply univocally (i.e. in the same sense) to an existing and a non-existing thing. This doctrine is also found in his later works.

According to Hackett, Bacon argues against the common teaching of the schools, particularly at Paris, according to which a word has a natural meaning from which, once given, it cannot fall away (cadere a sua significatione). On the Parisian view, the name 'Caesar' has the same meaning now that Caesar is dead, as it did when Caesar was living. Bacon argues that words can only stand for present things, and the being is not 'univocal': there is nothing in common between a being and a non-being, or between past, present and future things.

Thus Bacon rejects the predication of empty terms, and thus rejects propositions 'every man of necessity is an animal, with no men existing' (Omnis homo de necessitate est animal, homine non existente) and Caesar est homo, Caesare mortuo (Caesar is a man, with Caesar dead). Hackett says that Bacon's position was condemned by Robert Kilwardby in the Oxford condemnations of March 1277.

References

Braakhuis, H.A.G. in 'Kilwardby versus Bacon? The Contribution to the Discussion on Univocal Signification of Beings and Non-Beings Found in a Sophisma attributed to Rober Kilwardby', in Medieval Semantics and Metaphysics. Studies dedicated to L.M. De Rijk, PhD on the Occasion of his 60th birthday, ed. E.P. Bos, (Artistarium, Supplementa 2), Nijmegen, Ingenium Publishers, 1985, 126-142 [extracts].
Broadie, A., 'Robert Kilwardby', in A Companion to Philosophy in the Middle Ages, ed. Gracia & Noone, Oxford 2006.
De Libera, A., (1997) 'Roger Bacon et la logique', in J. Hackett, ed., Roger Bacon and the Sciences: Commemorative Essays (67-102), Leiden, New York and Cologne: Brill.
De Libera, Alain 'Roger Bacon et la référence vide. Sur quelques antécédents médiévaux du paradoxe de Meinong', Lectionum varietates, Hommage à Paul Vignaux (1904-1987) (Etudes de philosophie médiévale, LXV), Ed . J. Jolivet, Z. Kaluza, A. de Libera, Paris, Vrin, 1991, p. 85-120.
Ebbessen S. and Pinborg J., 'Studies in the Logical Writings attributed to Boethius of Dacia, 1970, (38-9).
Hackett, 'Roger Bacon', in A Companion to Philosophy in the Middle Ages ed. Gracia & Noone, Oxford 2006.
Henry, D.P. That Most Subtle Question, Manchester 1984. (esp. ch. 5 'Disputed Questions').
Kilwardby, Robert, 'Omnis homo de necessitate est animal' ed. S. Ebbesen & J. Pinborg, in 'Studies in the Logical Writings attributed to Boethius of Dacia', CIMAGL, 3 (1970), 37-40 [extracts].
Kretzmann, N., 'William of Sherwood's Introduction to Logic', University of Minnesota 1966.
Lewry, Osmund (1981), 'The Oxford condemnations of 1277 in grammar and logic', in H.A.G. Braakhuis, ed., Sprache und Erkenntnis im Mittelalter, Berlin, De Gruyter.
O'Donnell, J.R. (ed.), 'The Syncategoremata of William of Sherwood', Medieval Studies 3 (1941) p. 71. Paris Nat. lat. 16135, fol. 11rb-12vb
Ricardus Sophista, 'Omne coloratum est' ed. J. Pinborg, in Magister Abstractionum, CIMAGL, 18 (1976), 1-4.
Ricardus Sophista, 'Omnis homo de necessitate est animal' ed. ed. H.A.G. Braakhuis in 'English Tracts on Syncategorematic Terms &c', 161.
William of Sherwood, Introductiones in Logicam; ed. Ch. H. Lohr, P. Kunze & B. Mussler, in 'William of Sherwood, Introductiones in Logicam, Critical Text, Traditio, 39 (1983), 222-99.
William of Sherwood, Syncategoremata; ed. J.R.O'Donnell, C.S.B., in 'The Syncategoremata of William of Sherwood', Medieval Studies, 3 (1941), 46-93.

Edward Buckner. London, August 2007.

Latin	English
Ricardus Sophista	(Source: taken from Peter King's website – not attributed, probably Pinborg 1976)
OMNIS PHOENIX EST. Probatio: Haec est falsa, 'aliqua phoenix non est'; ergo eius contradictoria est vera, scilicet 'omnis phoenix est'.	Every phoenix exists [est]. Proof: 'Some phoenix does not exist' is false, therefore its contradictory, i.e. 'every phoenix exists' is true.
Sed contra: hoc signum 'omnis' exigit adminus tria appellata, et non sunt tres phoenices, ergo prima falsa.	But on the contrary side, the sign 'every' requires at least three appellata, and there are not three phoenixes, therefore the first is false.
Solutio: Prima est falsa eo quod affirmatur 'esse' de eo quod non est secundum actum, unde respondendum est per interemptionem, cum dicit hanc esse falsam, 'aliqua phoenix non est'. Est enim vera; vere enim negatur esse ab eo quod non est, et iste terminus 'phoenix' supponit pro non ente aequaliter ut pro ente.	Solution: The first is false in that being [esse] is affirmed of what does not exist in actuality, wherefore it is to be answered by interemption [i.e. by total destruction of the argument], when he says that 'some phoenix does not exist' is false, for it is true, for it truly denies being of that which does not exist, and the term 'phoenix' supposits for a non being equally as for a being.
Habita diversitate distinctionum adhibendarum, signo addito termino simplici, restat ut videatur diversitas sophismatum, signo universali addito termino composito.	According to the diversity held of the distinctions to be applied, when the [universal] sign is added to a simple term, it remains that there may seem to be a diversity of sophismata with the universal sign added to a composite term.
'Necessarium per se' est quod non potest nec potuit nec poterit esse falsum, ut hominem esse animal, et propter hoc dicitur quod haec est vera: OMNIS HOMO DE NECESSITATE EST ANIMAL. Quod sic probatur: Haec est vera et semper fuit vera et semper erit vera, 'omnis homo est animal', ergo necessaria; ergo omnis homo de necessitate est animal.	'Necessary through itself' is what cannot nor could be nor will be false, such as a man being an animal, and on account of this it is said that this - 'EVERY MAN OF NECESSITY IS AN ANIMAL' is true. Which is proved as follows: 'every man is an animal' is true, and always was true, and always will be true, therefore it is necessary, therefore every man of necessity is an animal.
Sed contra: Omnis homo de necessitate est animal, Sortes est homo, ergo Sortes de necessitate est animal.	On the contrary side: Every man of necessity is an animal, Socrates is a man, therefore Socrates of necessity is an animal. [which is absurd]
Solutio: Solet dici quod prima vera est et non valet: 'omnis homo de necessitate est animal, Sortes est homo, ergo Sortes de necessitate est animal'; commutatur enim unus modus supponendi in alium; ampliatur enim suppositio huius termini 'homo' in prima propositione universaliter unde habet modum indeterminate supponendi, et supponit determinate cum dicit 'Sortes est homo.' Unde est ibi fallacia figurae dictionis. Et alio modo est ibi fallacia aequivocationis, quia in prima propositione est esse habitudinis sive consequentiae et in secunda ut nunc, cum dicitur operatio entis, et sic aequivocatur esse. Sed modo videtur quod prima propositio sit falsa et non valet: 'haec est vera 'omnis homo est animal' et semper fuit vera et semper erit haec vera, ergo omnis homo de necessitate est animal', quamvis etiam dicatur quod non potest non esse vera.	Solution: It is customary to say that the first proposition [every man is an animal] is true, and 'every man of necessity is an animal, Socrates is a man, therefore Socrates of necessity is an animal' is not valid. For one manner of suppositing is changed into another, for the supposition of the term 'man' is 'ampliated' universally in the first proposition, wherefore it has a manner of suppositing indeterminately, and supposits determinately when we say 'Socrates is a man'. Wherefore it is, there, a fallacy of a figure of speech. And in the other way it is, there, a fallacy of equivocation, because in the first proposition it is the being of conditionality [habitudinis] or consequence, and in the second, of 'as now' [ut nunc], when it is called the operation of being, and thus 'being' is equivocal. But in this way it seems that the first proposition is false, and [the inference] 'every man is an animal' is true, and always was true, and always will be true, therefore it is necessary, therefore every man of necessity is an animal' is not valid, although also it may be said that it cannot not be true.
Sed sic sequitur quod haec propositio est necessaria, 'omnis homo est animal' et haec similiter, 'omnem hominem esse animal est necessarium', quae est multiplex secundum compositionem et divisionem, et vera in sensu compositionis, quia in hoc sensu significat quod haec propositio est necessaria, 'omnis homo est animal'; et in sensu divisionis falsa est, et ille sensus exprimitur per hanc, 'omnis homo, de necessitate est animal' Secundum unum sensum est enim multiplex ex eo quod modus potest determinare verbum ratione compositionis vel ratione rei verbi. Si ratione compositionis, significatur quod haec propositio est necessaria 'omnis homo est animal'; si ratione rei verbi, tunc significatur quod iste homo et ille et sic de singulis sit animal necessario, et hoc est falsum.	But it follows thus that the proposition 'every man is an animal' is necessary, and similarly 'that every man is an animal is necessary', which is manifold in respect of composition and division, and true in the sense of composition, because in this sense it signifies that the proposition 'every man is an animal' is necessary, and in the sense of division it is false, and that sense is explained by 'every man, of necessity, is an animal'. For according to one sense it is manifold in that the mode can determine the verb by reason of composition, or by reason of the thing corresponding to the verb. If by reason of composition, it is signified that the proposition 'every man is an animal' is necessary. If by reason of what corresponds to the verb, then it is signified that this man and that man and thus singly, is an animal necessarily, and this is false.
Pseudo-Kilwardby: Utrum haec sit vera, 'homo est animal', nullo homine existente	(Source: De Libera, loc. cit.) Whether 'a man is an animal' is true, even though no man exists
Sicut innuit Aristoteles in primo Peryermanias principio 'voces sunt notae sive signa intellectuum, intellectus vero sunt signa rerum'. Unde vox non est signum rei nisi per intellectum. Cum igitur possibile sit qualibet re destructa intellectum eius sive speciem in anima remanere et significari per vocem, manifestum quod nullo homine existente possibile est intellectum eius vel speciem in anima remanere et significari per vocem.	Just as Aristotle suggests at the beginning of the first book of the Perihermaneias, 'utterances are marks or signs of understandings, but understandings are signs of things'. Wherefore an utterance is not a sign of a thing except by understanding. Accordingly, since it is possible that with any thing destroyed, the understanding of it, or the species in the soul remains, and be signified by an utterance, it is manifest that with no man existing it is possible for the understanding of him, or a species in the soul, to remain and to be signified by an utterance.
Tunc arguo: Quae est comparatio hominis actualiter significati ut res ad animal ut res significatum, eadem est comparatio intellectus sive speciei hominis significati per vocem ad intellectum sive speciem animalis significatum per vocem. Sed naturalis est inhaerentia hominis ut res actu ad animal ut res actu, haec enim est in naturali materia 'homo est animal'. Ergo similiter est naturalis inhaerentia intellectus sive speciei hominis significati per vocem ad speciem animalis significatum per vocem. Ergo enuntiatio enuntians unum de altero erit vera, ut cum dicitur 'homo est animal'. Ergo haec est vera simpliciter 'homo est animal', nullo homine existente.	Then I argue: the relation [comparatio] between a man actually signified as a thing, to an animal as thing signified, is the same as the relation of the understanding or of the species of a man signified by an utterance to the understanding or species of an animal signified by an utterance. But the inherence of a man as thing in actuality, to an animal as thing in actuality is a natural one, for 'man is an animal' is in natural material. Therefore, the natural inherence of the understanding or species of a man signified by utterance, to the species of animal signified by utterance, is similar. Therefore, the assertion asserting one of the other will be true, such as when 'a man is an animal' is uttered. Therefore 'a man is an animal' is true without qualification, with no man existing.
Anon	(Source: De Libera)
Aliis videtur quod haec, etsi nullum suppositum hominis maneat, sit vera. Et isti dicunt quod non sequitur 'est animal, ergo est'; immo est paralogismum secundum quid et simpliciter, quia in prima copulat esse essentiae, in alia esse actuale. Et si arguatur eis quod tertium ponit secundum, ergo hoc verbum tertio adiacens ponit se secundo adiacens, quare sequitur 'est animal, ergo est', dicunt quod verum est ubi est esse idem commune ratione, scilicet, in utramque ponitur esse actuale, ut hic: 'Est albus, ergo est', &c. Ulterius. Sequitur 'est ens, ergo est'. Dicunt quod duplex est esse. Uno modo est esse essentiae, et tale est actus entis, ut est nomen; alio modo est esse existentiae, et tale est actus entis, ut est participium. Et ita nec unum esse infert aliud, nec unum ens aliud. Cum autem superadditur huic verbo 'est' secundum exigentiam subiecti, copulat tunc esse essentiae, tunc autem non, {sed} copulat esse actuale. Ideo non sequitur 'est ens, ergo est'; semper enim copula contrahitur per adiunctum.	To others it seems that this [proposition] is true, even if no suppositum of a man remains. And these people say that 'he is an animal, therefore he is' does not follow, indeed it is a paralogism [i.e. an argument that is unintentionally invalid] both in a qualified and unqualified sense, because in the first the copula represents [copulat] the being of essence, in the other, actual being. And if it is objected to them that the third gives the second, therefore the word ['is'] as a copula [tertium adiacens] presents itself as an absolute predicate [secundum adiacens], wherefore 'he is an animal, therefore he is' does follow, they say that it is true where being is the same, common in nature [ratione], namely, in both there is given actual being, such as here: 'it is white, therefore it is' &c. Furthermore, 'it is a being, therefore it is' follows. They say that being is of two sorts. In one way it is the being of essence, and such is the actuality of being, as it is a name, in another way it is being of existence, and such is the actuality of being, as it is a participle. And thus neither one being implies the other, nor one entity the other. But since the word 'is' is added according to the requirement of the subject, the copula then represents the being of essence, but when not, it represents actual being. For that reason 'it is a being, therefore it is' does not follow, for the copula is always contracted through an adjoined [term].
[...]
Nulla propositio verior est illa in qua idem praedicatur de se; sed in hanc: 'Omnis homo de necessitate est animal' praedicatur idem de se, quia genus de specie; genus autem est de significato speciei, et si animal non de necessitate sit, adhuc tamen de necessitate potest inesse et hoc inferiori suo, quia talis inhaerentia non ponit existentiam extremorum secundum esse actuale, sed secundum esse essentiae vel habitudinale.	No proposition is more true [than one] in which the same thing is predicated of itself [N1], but in [the proposition] 'Every man of necessity is an animal' the same thing is predicated of itself, because the genus [is predicated] of the species. But the genus is of the significate of the species, and if an animal does not exist of necessity, nevertheless still of necessity it can also be in its inferior, because such inherence does not suppose the existence of the terms according to actual being, but according to the being of essence, or 'conditional being'
Ibidem 102rb
Praeterea alii distinguunt istam 'omnis homo de necessitate est animal' eo quod hoc verbum 'est' potest dicere esse habituale vel esse actuale. Si dicat esse habituale dicunt quod est vera, si actuale falsa. Sed quia hoc verbum 'est' secundum adiacens semper copulat rem suam quae est existens, quando autem est tertium adiacens copulat rem illius cui apponitur, credo quod possit copulare esse habituale vel actuale, nunquam tamen in eadem propositione. Sed quandocumque praedicatur secundum adiacens copulat esse actuale, quandocumque autem praedicatur tertio adiacens cum praedicato essentiali copulat esse habituale, ut est 'copulare esse habituale' dicere extrema esse 'habitu vel in potentia', sicut quidam dicunt. Sed esse habituale appellabant antiqui dicere habitudinem vel inhaerentiam extremorum, et non existentiam eorum. Dicere autem esse actuale est dicere eorum existentiam. Dictum autem est quod hoc verbum 'est' semper dicit eorum existentiam ut quando copulat esse existentiae, quia existentia necessario ponit essentiam, sicut posterius suum prius, sed non convertitur, quia prius non ponit suum posterius.	Furthermore, others distinguish the proposition 'every man of necessity is an animal' in that the verb 'is' can bespeak conditional being or actual being. If it bespeaks conditional being they say that it is true, if actual [being], [it is] false. But because the verb 'is' as an absolute predicate [secundum adiacens] always joins its own significate [rem], which is existing, but when as a copula [tertium adiacens], it joins the significate of that thing to which it is appended, I believe that it could join conditional being, or actual being, yet never in the same proposition. But whenever it is predicated as an absolute predicate it joins actual being, and whenever it it is predicated as a copula, since by an essential predicate it joins conditional being, for example 'to join conditional being' means [dicere] the terms exist conditionally or potentially, just as certain persons say. But the ancients used the term [appellabant] 'conditional being' to mean the condition or inherence of the terms, and not their existence. But to say [dicere] actual being is to mean [dicere] their existence. But it was said that the verb 'is' always bespeaks their existence when it joins the being of existence, because existence necessarily supposes essence, just as what comes after [supposes] what came before it, but not conversely, because what comes before does not suppose what comes after it.
William Sherwood: Introductiones in Logicam	(Source: De Libera, op.cit., & Kretzmann 1966)
[Latin to follow]	[From Kretzmann 1966, pp 124-5]. The reason for adding the phrase 'having sufficient appellata' is that if [a common term] does not have sufficient appellata it can supposit for something that does not exist. And notice that sufficient appellata amount to at least three; therefore if there are not that many appellata a term can supposit for something that does not exist – e.g., if there are only two men 'every man exists' is false.
[Latin to follow]	[124] Proof: the sign 'every' is imposed for the purpose of distributing for the greatest plurality and the sign 'both' for the least. Therefore since the least plurality is in a pair, the sign 'both' distributes only for two. But the greater plurality is in a group of three or more; therefore the sign 'every' distributes for at least three [N2]. Therefore, since in the example above there are two supposita existing and the sentence means (et vult iste sermo) that the predicate is in [at least] three, it means that it is in something that does not exist. And so 'every man exists' is false and its contradictory, 'some man does not exist' is true. Therefore a man does not exist. But nonexistence is not predicated of something that exists; therefore 'man' supposits for something that does not exist.
[Latin to follow]	But [suppose someone says that] on the contrary 'every animal exists' is true and therefore 'every man exists' is true, for the term 'animal' has enough appellata.
[Latin to follow]	[In that case] we must point out that in the case [in which there are only two men] this does not follow: 'every animal exists; therefore every man exists', because the term 'animal' does have sufficient appellata, and so supposits only for what exists, while 'man' does not, and so supposits for something that does not exist. And by the same reasoning this does not hold: 'a man does not exist; therefore an animal does not exist'.
[Latin to follow]	But again, consider this counterargument: 'every animal is, every man is an animal; therefore every man is [N3]' The first premiss is true, and the second is necessary, since the genus is predicated of a species.
[Latin to follow] Cum autem dico: Omne animal est, praedicatur hic esse actuale, quod est existere. Cum autem dico: Omnis homo est animal, praedicatur esse habituale. Et secundum quod necessaria est, habet virtutem huius conditionalis: Si homo est, animal est. Cum enim est ponitur inter haec extrema: homo, animal, tamquam medium, dicit habitudinem mediam inter haec duo.	[In response to this] we must point out that the argument is not valid [non valet]. [Kretzmann ends]. For when I say 'every animal is', actual being is predicated here, which is to exist. But when I say 'Every man is an animal', conditional being is predicated. And according as it is necessary, it has the force of the conditional 'if a man is [i.e. exists], an animal is'. For when 'is' is put between the terms 'man' and 'animal', as if an intermediary, it bespeaks a conditional relation between the two. (check this in the translation).
William Sherwood: Syncategoremata	(source: De Libera, from O'Donnell 1941)
Sciendum autem quod hoc verbum 'est' quandoque accipitur aequivoce; dicit enim quandoque esse actuale quod debetur actuali existenti, quandoque esse habituale quod debetur ei quod in se est natura aliqua et natum est habitualiter esse in aliquo singulari, licet non actualiter sit. Primo modo haec est falsa: Omnis homo est animal, nullo homine existente. Secundo modo vera.	You must know that the word 'is', is sometimes taken equivocally, for it sometimes bespeaks actual being, which is owed to actual existence, sometimes conditional being, is owed to what in itself is a certain nature, and which is suited conditionally to be in some singular thing, although it is not actually in one. In the first way, the [proposition] 'Every man is an animal' is false, with no man existing. In the second way, it is true.
Roger Bacon	(source: De Libera, op. cit.)
Probant quod vox non potest cadere a sua significatione, sed hoc supponunt tamquam rationem infinitorum quae existimant verissima cum sint falsissima, ut quod homo est animal nullo existente, et quod Caesar sit homo [...] et alia innumerabilia erronea, sicut ex praecedenti ratione sua falsissima quod nomen significat aliquid commune enti et non enti eliciunt mendacia paene infinita.	They 'prove' that an utterance cannot fall away from its signification, but they suppose this as if it were the reasoning of infinitely many things which they judge most true, but which are most false, such as that a man is an animal, with no [man] existing, and that Caesar is a man […] and other innumerable errors, just as from the preceding argument at its falsest, that a name signifies something to a being and non a non being, they spin out lies almost infinite [in number].

[N1] This is apparently said by Boethius, and is mentioned by many other writes, including Ockham. So far, I have not been able to locate it in Boethius.
[N2] Kretzmann: cf Sherwoods' Syncategoremata (O'Donnell, pp. 49-50).
[N3] Kretzmann notes that omne animal est and omnis homo est which are elsewhere translated by him as 'every man exists' &c are translated using 'is' rather than 'exists', to suit the example.

THE LOGIC MUSEUM Copyright (Introduction and original translations) C E.D.Buckner 2007.