![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
From the nineteenth century onwards, the so-called ‘Debate over Universals’ from the time of Peter Abelard (c. 1100–50) is regarded as one of the major problems of medieval philosophy. This paper tries to advance the current approaches (which fall into two broad categories: a more analytical line of enquiry and a more historical one) by establishing a new understanding based on assumptions and questions. Through a comprehensive analysis of both published and unpublished logical commentaries from the first half of the twelfth century, it shows that – contrary to the received narrative – a lot is not discussed, but rather assumed. It therefore investigates what the debate is not about – arguing that this is just as relevant as what the debate is about.
Acknowledgements
I am immensely grateful to Irene Binini, Laurent Cesalli, Enrico Donato, John Marenbon, and the anonymous referees for the BJHP for their insightful remarks on earlier versions.
Notes
1 For instance, several collections of translated texts exist: Maioli, Gli universali; Wöhler, Texte zum Universalienstreit; Spade, Five Texts.
2 As evidenced by, among others, a famous passage of Abelard’s Historia Calamitatum, §6 (Luscombe, The Letter Collection, 8–11), and John of Salisbury’s Metalogicon, II, 17.
3 See, for instance, Cousin, Ouvrages inédits d’Abélard. Abelard had been connected to the debate over universals well before: on all this, see de Libera, L’archéologie philosophique, 125–76 and, more generally, Jolivet, “Les études”; and Santinello and Piaia, Storia.
4 See de Libera, La querelle, 11.
5 See Cousin, Ouvrages inédits d’Abélard, lxv.
6 See especially König-Pralong, Médiévisme, 131–63.
7 This portrait, it should be noted, also owes a lot to Abelard’s Historia calamitatum.
8 Now edited in King, “Pseudo-Joscelin”.
9 See King, “Why” and, on the “non-problem” of substance dualism for twelfth-century cosmologists, Hicks, Composing the World, 113–50.
10 Galluzzo, “The Problem”, 345; cf. 343, 350; see also Panaccio, “Universals”, and Pinzani, The Problem.
11 See de Libera, La querelle, 17; L’archéologie philosophique; and, among other contributions, “Archéologie et reconstruction”.
12 I use here a common-sense notion of both question and answer, although several twelfth-century authors (and their sources) explicitly discuss what a question is. Claude Panaccio also puts forward a distinction between questions and problems in his Récit et reconstruction, 100–14.
13 See also de Libera, “Archéologie et reconstruction”, 565–8.
14 See also Chiaradonna and Galluzzo (eds), Universals in Ancient Philosophy; and Gerson, “Platonism”. Gerson claims that the Problem of Universals came to be “badly formulated” and there was a “gradual mutation of the problem” from antiquity to Porphyry; see also below, section 3. For a contemporary approach, see Rodriguez-Pereyra, “What is”, and Galluzzo and Loux (eds): The Problem.
15 Considerations similar to the above, it seems, led John Marenbon to distinguish in re/in voce exegesis of certain authoritative passages from statements committing to real universals or to a particularist ontology (“Life, milieu”, 27–34; and “Logic at the Turn”, 191–2). Not every twelfth-century statement about genera and species enters the so-called Debate over Universals: see Marenbon, “Logic at the Turn”, 192. The above distinction has later been reconsidered, and he now prefers to talk about the development of a “language-focused logic” – concerned with words that refer to words, rather than words that refer to things; see “Logic at the Turn”, 201–15.
16 For example, Anne Grondeux and Irène Rosier-Catach have unearthed a fascinating discussion as to whether numerus (number) is a genus, of which binarius (two/couple) would be a species and hic binarius (this couple) an individual. They have also alerted their readers to several other discussions of collections relating to conjugation and collective names; see Grondeux, Rosier-Catach, Priscien lu, vol. 1, 145–80. In the discussion over number, certain truths about genera and species are assumed, and it is verified whether numerus and binarius comply with them (with several texts reaching the conclusion that this is not, in fact, the case). In other terms, the discussion over number should not be identified with the Problem of Universals, nor with discussions of collectio-theories over universals, although it is a discussion of collections (number is a collectio unitatum). Similarly, I would not describe the problem of the unity of a proferred utterance (vox) in the ears of its many listeners as another form of the Problem of Universals (as one reads in Rosier-Catach, “Vox”, 105). The proferred vox is always regarded as an individual, although, as for universals, questions are raised concerning its unity and identity in many. See also Tarlazzi, Individui universali, 321.
17 This choice of sources is, one might complain, quite traditional. Indeed, once the different questions addressed by this sort of sources have emerged (and this is my purpose here), it will be clear that relevant remarks for some of them can be found elsewhere, too. For instance, remarks on what genera and species are (one of the ‘core’ questions) can be found in commentaries on Aristotle’s Categories and De interpretatione and on Boethius’ De divisione and De differentiis topicis. In turn, detailed discussions of the semantics of common names (one of the ‘ancillary’ questions) are also to be found in commentaries on Priscian’s Institutiones; see Rosier-Catach, “Les Glosulae”. Other types of sources are also emerging, such as commentaries on Boethius’ Consolatio and Opuscula sacra.
18 In his personal database, Yukio Iwakuma catalogues them as U1/De generibus et speciebus (henceforth GS), edited in King, “Pseudo-Joscelin”; U2/‘Quoniam de generali’ (QG), edited in Dijs, “Two Anonymous”, 93–113; U3/Sententia de universalibus secundum magistrum R., edited in Dijs, “Two Anonymous”, 113–17; U4/P21 (see below); and the short U5 in München, Bayerische Staatsbibliothek, Clm 29520(2), f. 1ra–vb. I have also considered independent logical treatises, such as Abelard’s Dialectica and Garlandus’ Dialectica. See Appendix for more details on the formulations of questions.
19 For Isagoge commentaries (P1, P2 etc.), I have considered the surviving commentaries listed in Marenbon, “The Isagoge” (updating previous versions of the catalogue).
20 Both edited in Boethius, In Isagogen (henceforth 1Isag. and 2Isag.). In the High Middle Ages, 2Isag. was disproportionately more studied and copied than 1Isag. Some twelfth-century commentaries, however, do take 1Isag. into account: for instance, P2, P13, P13a/P33, and P16.
21 See also Sorabji, “Universals”.
22 The distinction emerges in e.g. LI 23.2–4; Garlandus, Dialectica, 12.20–23. For a sort of counter-example, mentioning ensis (sword) as a species and instrumentum (tool) as a genus, see Boethius’ 2Isag., 192.13–16 (inspiring passages in P3, P15, P16, and P20).
23 I leave aside the complex issue of differentiae. They are predicated in quale, but enter into the definition, that is, capture the essential features of that of which they are predicated.
24 A further distinction should be made between genera and species. For instance, P20 states that a genus is the “matter” of a certain thing (materia rei), whilst a species is the “whole being” of that thing (totum esse rei); see Wien, Österreichische Nationalbibliothek, VPL 2486, f. 48va; a similar distinction is to be found in P3, version A (Iwakuma, “Pseudo-Rabanus”, 171). But differences tend to be glossed over, especially if compared to ancient sources; see Brumberg, “Le problème”, 78–80, and Erismann, L’Homme, 89. For instance, it is common to qualify species as matter, as one reads in GS §§101–4.
25 This section takes inspiration from, and expands on, the principles identified in Marenbon, The Philosophy, 104, 117–23; see also Erismann, L’Homme, 73–148.
26 If questions are occasionally raised, the answer is straightforward. For instance Boethius mentions, against assumption 5, the possibility that universal concepts be empty (see n. 54). He replies in the negative and, to the best of my knowledge, universal concepts are never positively argued to be empty in the twelfth century. There are also some doubtful cases (see n. 45) which highlight, by contrast, the assumption in place in most cases.
27 See Marenbon, The Philosophy, 104: ‘naturalism’ is the claim that “all particular substances belong to natural kinds, which are identified according to the scheme of genera, species, and differentiae” (and see also Marenbon, The Philosophy, 368).
28 See below, assumption 4.
29 See Erismann, L’Homme, 81. This is true even when a species comprises one individual only.
30 In L’essentialisme, 28–32, Magali Roques lists three different ways in which ‘essentialism’ is used by medievalists today: for metaphysical theories, inspired by Avicenna, endorsing common natures indifferent to existence; in modal logic, for theories endorsing de re modalities; for the role played by quiddites in commentaries on Aristotle’s Metaphysics for justifying the identity and unity of a certain thing. I use the term here in a sense which is – I believe – a simplified version of the last. Although key Aristotelian sources were not available, the link between universals and definitions is clearly established in e.g. Boethius, 1Isag., 10.25–11.23, 18.1–19, 19.19–21.13, 42.8–44.5, 59.12–63.16, 127.10–22; 2Isag. 153.8–154.2, 155.19–156.9, 180.21–181.8, 196.18–197.12, 202.1–203.4, 222.11–223.24, 246.18–248.10, 273.9–275.2. This use of ‘essentialism’ is only partially linked to the twelfth-century use of the word essentia, which may, but does not always, mean essence (see Jolivet, “Notes”).
31 By comparison, Locke’s distinction between real and nominal essences questions the assumptions of essentialism and strong naturalism; see as a starting point, Galluzzo, Breve storia, 165–70.
32 See Erismann, L’Homme, 36, 39–60, 78, 137–9.
33 Petrus Abaelardus, Dialectica, 283.12; see also 280.13 (lex naturae); Jolivet, “Non-réalisme”, 193–4. Abelard also claims that hypothetical propositions expressing rerum habitudines, namely, how things are arranged (such as “if it is a man, it is an animal”) are true ab aeterno: see Petrus Abaelardus, Dialectica, 279.10–283.35. Prior to such hypotheticals are some categorical propositions (e.g. for the hypothetical above, “animal is the genus of man”) that show the nature of things (rerum natura) and “preserve for ever the necessity of the consecution”: see Petrus Abaelardus, Dialectica, 283.11–19 and Martin, “Logic”, 180–1, 183–7.
34 See also LI 16.22–18.3, distinguishing between the points of view of grammarians and dialecticians. Grammatical construction (constructio) simply assures that a statement is well-formed, irrespective of whether things are as the statement says they are. Dialectical predication (praedicatio), in contrast, “pertains to the nature of things and to showing the truth of the state of things” (ad rerum naturam pertinet et ad ueritatem status earum demonstrandam). ‘Stone’ can be construed with ‘man’ in a well-formed statement (‘Man is a stone’) but it cannot, properly speaking, be predicated of ‘man’. See Jolivet, “Non-réalisme”, 183 and 185.
35 See Cameron, “The Constraints” and Binini, “Some further remarks”.
36 See LI 19.21–20.14; 23.2–12, and Jolivet, “Non-réalisme”, 194–5; Marenbon, “Abelard’s Theory”, 44–8.
37 See e.g. Plato, Meno, 72a–d.
38 Chiaradonna and Galluzzo, “ Introduction”, 3.
39 Twelfth-century century authors derive their list of essential properties from authoritative texts, and occasionally raise issues with them. For instance, in LI 90.31–91.6, Abelard is puzzled by the fact that gressibilitas (the ability to walk) and bipedalitas (being two-footed) are classified by authorities as substantial differences, while risibilitas (being capable of laughter) is a proprium, and therefore an accident. Elsewhere, he discusses at length how bipedalitas enters into the definition of a human being, although some men are amputees (LI, 104.18–105.38, building on Porphyrius, Isagoge, tr. Boethius, 28.13–17 and Boethius, 2Isag., 330.113–331.9); see also Martin, “Abelard”, 99–113.
40 See GS §§90, 113, 123, 138, 199 and below, n. 49.
41 See e.g. QG §§2–5; GS §33; LNPS 515.14–31.
42 Marenbon, “Abelard on Differentiae”, 190, reconsidering Marenbon, The Philosophy, 175, 195–201.
43 See Marenbon, The Philosophy, 117.
44 See Marenbon, The Philosophy, 104: ‘strong naturalism’ is naturalism, plus the claim that “at least in principle, we unproblematically identify the members of each kind”; see also Marenbon, The Philosophy, 368.
45 See the excellent reconstruction in Rosier-Catach, “Vox”. It is debated whether vox is a substance, a quality, a quantity, or belongs to no category at all. Interestingly, the discussion originates from conflicting authorities (Aristotle, Priscian, and Boethius).
46 Abelard comments on our ignorance in this respect, as highligted by Marenbon, The Philosophy, 119, 236; see also P3 in Iwakuma, “Pseudo-Rabanus”, 94, on the unknown difference of donkey.
47 Others have been mentioned above: natura, status, the eternal truth of certain categorical and hypothetical propositions, the difference between the points of view of grammarians and dialecticians.
48 See, e.g. LNPS 558.28–559.4.
49 I borrow this terminology from Arlig’s remarkable discussion in A Study, 278–80. Although his approach is rather different from mine, Arlig concludes that classes are built on resemblance, and resemblance is grounded in similarity in creation (see A Study, 290–5, and above, n. 40). In “Introduction”, 15–16 n. 28 Arlig notes that “a major problem for collectio-theories” is that “they need to provide an answer to this question: What is the reason why all these concrete individuals, these and none other, constitute the universal?”, and then mentions a trope theory resting on divine ideas as the solution of GS. My approach is rather to claim that, because of their assumptions, twelfth-century theorists were not trying to answer Arlig’s question. Grouped things, and definitions given by authorities, are their starting point.
50 See LI 15.1–15.
51 See LI 14.40–15.1.
52 See especially de Libera’s discussion of fictitious concept (ψιλὴ ἐπίνοια), embedded into Porphyry’s questionnaire: Porphyrius, Isagoge, tr. de Libera, vii–cxlii (but cf. tr. Barnes, 37–49). Non-emptiness of generic and specific concepts, one might wish to say, was not an assumption for Porphyry (and Boethius), but the object of debate against the Stoic identification of universals with imaginary fabrications; see de Libera, L’archéologie philosophique, 78–80.
53 In De intellectibus §59 Abelard states that the understanding generated by ‘homo’ would be empty were there absolutely no men existing – a conditional sentence with imperfect subjunctives, implying conditions contrary to fact.
54 This issue is also addressed in Boethius, 2Isag., 160.6–10, and 163.6–167.17; see above n. 46 and, for a commentary, among others Marenbon, “Logic before 1100”, 12–14. As Marenbon highlights, Boethius holds that “genera and species are not corresponding thoughts”, but denies that “thoughts that are not corresponding are empty and false”. Twelfth-century discussions take Boethius’ analysis as their starting point. In LI 18.7–19.6 Abelard states that there is a doubt on the signification of universal words, for they do not seem to have a res subiecta, and do not – it seems – generate a firm (sanus) understanding about something. He will then explain that “this is not so” (19.7); universal words have a signification about individual things by nomination (19.7–20.16) and universal concepts are abstract, but not empty (cassi) (24.38–27.34); see also LNPS 524.35–525.4 and 530.20–531.29.
55 See Erismann, L’Homme, 37–8, 195, 294.
56 See Erismann, L’Homme, 81–2, 84, 89–90, 101, 268, 300.
57 Realism is also supported on authoritative ground, namely, by referring to passages from authorities; see e.g. QG §6; LI 9.21–28; LNPS 512.3–513.14.
58 See Galluzzo, Breve storia, 58–61. Galluzzo’s ontological argument differs from my constituent argument. It concerns attribute agreement: universals – the argument goes – are needed to explain why two or more things have the same property.
59 Hollick, Anonymi Epternacensis, 341: “Subsistunt igitur genera et species. Nam si non subsistunt, nec indiuidua. Si nec indiuidua, nec nos qui loquimur. Sed hoc perspicue falsum est. Subsistunt igitur”.
60 QG §26: “Unde si genera et species sint – sunt autem, quippe materia indiuiduorum sunt – oportet quod indiuidua sint”. See also Erfurt, Wissenschaftliche Allgemein-Bibliothek, Amplon. Octavo 5, f. 3v (P13): “Si omnis res quae uere sunt sine his quinque esse non possunt, has ipsas quinque uere et non false esse necesse est; omnis autem res quae uere sunt, sine his esse non possunt; igitur uere subsistunt” (for italicised words, cf. Boethius, 1Isag., 26.1–2).
61 Cf. Porphyrius, Isagoge, tr. Barnes, 11.13–18 and 15.7.
62 Sources have complex features, which cannot be addressed here in detail. In addition to U-treatises, Marenbon lists thirty-three P-commentaries, but some no longer survive, and some are very incomplete.
63 Porphyrius, Isagoge, tr. Boethius, 5.10–14: “Mox de generibus et speciebus illud quidem siue subsistunt siue in solis nudis purisque intellectibus posita sunt siue subsistentia corporalia sunt an incorporalia, et utrum separata an in sensibilibus et circa ea constantia, dicere recusabo”; my translation modifies King, “Abelard’s Answers”, 249; see also below, n. 74.
64 See especially Porphyrius, Isagoge, tr. de Libera, xxxvi–lxxv; de Libera, L’archéologie philosophique, 78–87.
65 See Appendix for examples.
66 This question overlaps with two further discussions in Isagoge commentaries. One, addressed in the prologue, is the issue of what the Isagoge is about and Porphyry’s intention in writing it: see, for instance P3 (version P, Iwakuma, “Pseudo-Rabanus”, 61–2), and Sankt Gallen, Stiftsbibliothek, 134, p. 77a; P4 (Iwakuma, “Vocales”, 103–6); P14 in Paris, BnF, lat. 17813, f. 1ra–b; P16 in München, Bayerische Staatsbibliothek, Clm 14458, f. 83ra–b; P32 in Cashel, Bolton Library, 1, p. 118. The other, addressed in the section on genus, is the issue of what the definition of genus applies to; see e.g. Iwakuma, “Pseudo-Rabanus”, 95–8; P16 in München, Bayerische Staatsbibliothek, Clm 14458, f. 87ra–b; and P28 in Sankt Gallen, Stiftsbibliothek, 833, p. 4.
67 For instance, Porphyry’s first question is to be understood as two separate questions in Abelard’s interpretation, namely: (1) Do genera and species exist, or not? (2) Are genera and species posited in understandings that are alone, bare, and pure, or not?; see LI 28.11–15; de Libera, L’art, 473–6; King, “Abelard’s Answers”, and Panaccio, “Les huit questions”. In contrast, as Appendix shows, several sources reduce Porphyry’s first question to asking whether genera and species exist or not, wiping out concepts; e.g. P13, P15, P24, and Hollick, Anonymi Epternacensis, 340.
68 LNPS 512.19–515.9 lists authorities for the views that universals are (a) res, (b) intellectus, or (c) sermones. But (b) is not discussed further in 515.10–524.24, where (a) and (c) are analysed. See also Tarlazzi, Individui universali, 89–91.
69 The threefold distinction, Alain de Libera has shown, casts over the Isagoge a scheme from Aristotle’s Peri Hermeneias, and used in late antiquity for discussing the subject matter of Aristotle’s Categories. The distiction was already circulating before Cousin famously used it for describing Abelard as a conceptualist; see de Libera, L’archéologie philosophique, 39–42, 50–6, 151–76.
70 See also Iwakuma, “Vocales revisited”, 95, 97–8, 130, 166, and Grondeux, Rosier-Catach, Priscien lu, vol. 1, 124–30.
71 Several twelfth-century Isagoge commentaries (for instance, P3) contain no extensive discussion of the signification of universal words and of their understandings.
72 See Appendix for the Latin formulation.
73 See LI 27.35–30.5; King, “Abelard’s Answers”, 250; and de Libera, L’art, 281–498.
74 This depends on Abelard’s construal of the words ‘et circa ea constantia’; see de Libera, L’art, 299–301, 482–4; and King, “Abelard’s Answers”.
75 See LI 8.11–22 (quoted in Appendix); cf. 19.14–20 and, for replies, 19.21–24.37; see also LNPS 528.
76 See Galluzzo, “The Problem”, 336, 352.
77 See Cesalli, “L’universalité”. The fourteenth-century discussion – Cesalli argues – is best described as revolving around the question, ‘What is sufficient to explain universality, understood as a conceptual and linguistic fact?’.