The use of the title Ὄργανον to refer to Aristotle’s treatises on logic does not go back to the philosopher himself.Footnote ¹ At most, he observes in the Topics (8.14 163b9–11) that ‘the ability to see and comprehend the consequences of each of two hypotheses is no mean instrument (μικρὸν ὄργανον)’ for the purposes of knowledge and philosophical investigation. The earliest attestation of this noun as an explicit designation for what we call logic is found at the beginning of the third century a.d. in Alexander of Aphrodisias’ commentary on a different passage of the Topics (1.11 104b1–12). Alexander (In Top. 74.29) remarks that ‘within philosophy the study of logic has the place of an instrument’ (ὀργάνου χώραν ἔχει ἐν φιλοσοφίᾳ) and derives this idea from Aristotle’s own statement that dialectical investigations may achieve their goals (choice and avoidance or truth and knowledge) not only by themselves but also as a support (συνεργόν) for other investigations (104b2–3).Footnote ² Alexander draws a connection between this notion and Aristotle’s identification of ‘logical’, as opposed to ethical and physical, propositions (λογικαὶ προτάσεις, Top. 1.14 105b24), and surmises on these grounds that τὸ λογικὸν ὀργανικόν ἐστι (In Top. 94.8).Footnote ³

Much as they may make sense to a modern reader, Alexander’s remarks distort what Aristotle means by λογικός. Rather than pointing to what we would call ‘logical’ (which is closer to the Aristotelian meaning of ἀναλυτικός), this adjective indicates what is ‘verbal’ or ‘verbalizable’, hence ‘general’. What Alexander might have had in mind is the Hellenistic meaning of λογικός—in particular, the Stoic one (as in Diog. Laert. 7.39), which is indeed closer to the modern notion of ‘logic’.Footnote ⁴

Alexander’s remarks reflect a polemic between the Peripatetic and the Stoic schools regarding the nature of logic as a μέρος or an ὄργανον of philosophy—a debate which Alexander himself describes at the beginning of his commentary on the Prior Analytics (1.3–9). Unlike later sources, he does not mention the Stoics specifically (in fact, the bipartition of philosophy into theoretical and practical instead of a tripartition into logic, physics and ethics set the Peripatetics apart from most other schools).Footnote ⁵ Incidentally, later sources will present the Platonists as holding the middle ground, with logic being considered both an instrument and a part of philosophy.Footnote ⁶

If logic were (or at least could be considered) an instrument, it was only natural that treatises dealing with its aspects could be referred to as ‘instrumental’ (ὀργανικά). In the fifth century a.d., for example, Ammonius writes that, of the acroamatic works of Aristotle, τὰ μέν ἐστι θεωρητικὰ τὰ δὲ πρακτικὰ τὰ δὲ ὀργανικά (In Cat. 4.29). Unlike Aristotle’s own labels Τοπικά (An. pr. 1.1 24b12, etc.), Ἀναλυτικά (Eth. Nic. 6.3 1139b27, etc.) or Ἠθικά (Pol. 1.2.5 1261a31, etc.), which refer to the topic of each treatise, Ammonius’ designations are broad indications of the curricular function of the books rather than bibliographical references, so to speak. The closest we get to the use of ὄργανον as a title is perhaps a sentence from Ammonius’ commentary on Porphyry’s Isagoge—a text which became the standard introduction to Aristotle’s Organon in the manuscript traditionFootnote ⁷ —stating that ‘the present book is subsumed beneath the logical instrument of philosophy’ (ὑπὸ δὲ τὸ λογικὸν ὄργανον ἀνάγεται τῆς φιλοσοφίας τὸ προκείμενον βιβλίον, In Porph. Is. 23.19–20).Footnote ⁸ Still, τὸ λογικὸν ὄργανον is not the designation of a βιβλίον as such, much as the grouping of the treatises that we call the Organon—albeit with some fluidity—must have been accomplished in antiquityFootnote ⁹ and the idea that they constituted ‘a systematic whole’Footnote ¹⁰ goes back to at least as early as Porphyry (In Cat. 56.23–57.33).Footnote ¹¹

Nevertheless, a number of modern scholars are inclined to regard these passages as evidence for the early establishment of Organon as a collective title for what had come to be the introductory section of the Aristotelian corpus. RossFootnote ¹² and Reale,Footnote ¹³ for example, assert that this bibliographical designation goes back at least to the sixth century—in the nineteenth century some scholars would even project it as far back as to Andronicus.Footnote ¹⁴ Other scholars, however, are more cautious as to the antiquity of the name of the Organon. Gottschalk, for instance, observes that ancient commentators may well ‘describe logic as an organon of philosophy and logical writings as organika, but the use of “Organon” as a collective title for Aristotle’s logical works seems to have no ancient authority’.Footnote ¹⁵

This sceptical position invites the question of when Organon started being used unequivocally as a bibliographic designation. In this connection, in the nineteenth century Barthélemy Saint-Hilaire put forward the view that the logical treatises of Aristotle began to be commonly referred to as Organon no sooner than among fifteenth-century Latin commentators,Footnote ¹⁶ given that—as indeed seems to be the case—this title never recurs in the works of the likes of Psellus, Nicephorus Blemmydes or Gregorius aneponymus, who prefer the designation ἡ λογική (ἐπιστήμη, πραγματεία).Footnote ¹⁷ At least some Byzantine evidence for the use of Organon as a title, however, is hidden in plain sight.

To begin with, the editio princeps of Aristotle (the 1495 Aldine) opens with the following hexameters under the title εἰς ὄργανον Ἀριστοτέλους· ἀνώνυμον:Footnote ¹⁸

In the letter on the verso of the page containing the epigram, Manutius presents the epigram as evidence for the fact that ‘logici ac dialectici libri … Aristotelis organon graeci appellant’, mentioning that he found the poem ‘in antiquo codice’. The text printed by Manutius reads ἐκάλεσαν at line 2—which spoils the metre—and corresponds to the text of MS Vat. Barb. gr. 139 (fifteenth century), fol. 21v, which contains the epigram (right after the colophon located at the end of Categories and before the beginning of De interpretatione) under the same title and with the same layout as the Aldine.Footnote ¹⁹ This poem was reprinted in the front matter of Paci’s editions of the Organon (1584 and 1605)Footnote ²⁰ without the metrical error and with two different Latin metrical translations.Footnote ²¹

Unlike Aldus, we now know from MS Vat. Reg. gr. 116 (fourteenth century), fol. 2r that the epigram is probably the work of the fourteenth-century monk Isaac Argyros,Footnote ²² who compiled a manuscript containing Aristotelian works and commentaries (Neap. III D 37) in Constantinople.Footnote ²³ In the same century, we encounter ὄργανον as a bibliographical designation in a list of commentaries on the logical works of Aristotle (MS Marc. gr. Z 203, fol. 228v), which are presented under the header εἰς τὸ ὄργανον,Footnote ²⁴ and possibly in the rubrics of George Pachymeres’s (1242–1310) commentary on the Organon (MSS ÖNB Phil. gr. 150, fol. 1r and Vat. gr. 321, fol. 5v).Footnote ²⁵ The title Organon must thus have been current in the Byzantine scholarly world at least a century before Manutius. If we follow the lead of epigrams, however, we can go further back in time. The following dodecasyllables (In Aristotelis Organum)Footnote ²⁶ appear at the end of a late thirteenth-century manuscript (Laur. Plut. 72.3, fol. 147r):Footnote ²⁷

In the same century, the beginning of a colophon following the Sophistical Refutations in MS Esc. Σ.III.9, fol. 152r reads τέλος σὺν θεῷ τῶν σοφιστικῶν ἐλέγχων ἀριστοτέλους. ἴσως δ’ εἰπεῖν καὶ αὐτοῦ ὀργάνου.Footnote ²⁸ One can also find a reference to the Organon as a book in the so-called Anonymus Coislinianus, a commentary on De interpretatione whose oldest manuscripts (BNF Coislin 160 and Laur. 72.1) date to the thirteenth century and which presupposes Stephanus’s (turn-of-the) seventh-century commentary.Footnote ²⁹ This text presents the structure of τὸ πᾶν ὄργανον as being articulated εἰς τὰ πρὸ τῶν ἀναλυτικῶν καὶ εἰς αὐτὰ τὰ ἀναλυτικὰ καὶ εἰς τὰ μετὰ τὰ ἀναλυτικά before detailing which texts constitute each part (Prooemium, 5 Krewet).

With the current dating of Leon Magentenus to the twelfth century (before 1185),Footnote ³⁰ we can push the beginning of our evidence further into the past. In his commentary on the Posterior Analytics, Magentenus mentions that λέγεται δὲ ἡ ἀποδεικτικὴ ἐπιστήμη ὄργανον, ἐξ ἧς καὶ τὸ ὅλον βιβλίον Ὄργανον ἐκλήθη.Footnote ³¹ This (then unattributed) statement was interpreted by MielachFootnote ³² (followed by Waitz)Footnote ³³ as indicating that the Posterior Analytics (which Ammonius calls τὰ ἀποδεικτικά, In Int. 102.21; cf. Philoponus, In An. pr. 4.14)Footnote ³⁴ was the first book to be referred to as Organon, and that the title was then extended to the rest. This testimony is ambiguous, however, as to whether it was the treatise itself (as a bibliographical entity) or merely its content to be referred to as ὄργανον, and reveals nothing about when this designation was current.

The final candidate I was able to trace for the earliest use of Organon strictly as bibliographical designation occurs in a didactic ‘pseudo-dialogue’Footnote ³⁵ which is known as the Byzantinisches Schulgespräch Footnote ³⁶ and whose date might range between the mid eleventh and the early fourteenth centuries.Footnote ³⁷ Towards the end of the text (lines 68–72 Treu), the question is asked πόσα τῷ Ἀριστοτελικῷ ὀργάνῳ ἐμπεριέχεται, and the answer provided consists of a detailed list of the works constituting the Organon. The list consists of the following items: (1) κατηγορίαι, (2) περὶ ἑρμηνείας, (3a) τρία σχήματα καὶ πρότερα ἀναλυτικά (cf. Philoponus, In An. pr. 1.4), (3b) μίξεις (from An. pr. 1.8 29b29; cf. Philoponus, In An. pr. 119.1), (3c) περὶ εὐπορίας προτάσεων (from An. pr. 1.27 43a20; cf. Philoponus, In An. pr. 270.1 and Ammonius, In Int. 102.7), (3d) περὶ ἀναλύσεως συλλογισμῶν (from An. pr. 1.32 46b40; cf. Philoponus, In An. pr. 315.1), (4) ἀποδεικτική· τὰ δεύτερα τῶν πρώτων, (5) τοπικά, ἃ καλοῦνται καὶ διαλεκτικά (cf. Philoponus, In An. pr. 4.15), (6) καὶ σοφιστικοὶ ἔλεγχοι. As one can see, Porphyry’s Isagoge is omitted and the Prior Analytics are unpacked into four sections following a tradition established by Philoponus and adopted by later commentators.Footnote ³⁸

The list corresponds almost verbatim to that of the Anecdoton Hierosolymitanum,Footnote ³⁹ a book catalogue transmitted in a late thirteenth-century manuscript (which still lists the books of the Organon under the header ἡ λογικὴ πραγματεία rather than under ὄργανον). It is also reminiscent of the introductory section of the so-called Anonymous Heiberg,Footnote ⁴⁰ a compendium of logic and the quadrivium entitled συνοπτικὸν σύνταγμα φιλοσοφίας in MS Heidelberg Pal. graec. 281 (compiled in 1040), fol. 1r, which dates to 1007.Footnote ⁴¹ The beginning of this text reads:

The first three items in this list offer an almost exact match to those in the Schulgespräch, even though the compendium covers the topics of the Isagoge and skips those of De interpretatione. Footnote ⁴² At the same time, the only item which, strictly speaking, is an unequivocal bibliographical designation is τὸ περὶ ἑρμηνείας. Unlike κατηγορίαι, which could function both as the title and as a reference to the content of the book as a topic in the curriculum, and unlike σχήματα, which referred to the content and at most to the title of a section identified in the exegetical tradition, περὶ ἑρμηνείας was the traditional title of the treatise and a notoriously inadequate description of its content. The expression τὸ τῆς φιλοσοφίας ὄργανον—which will still recur in the thirteenth-century ὄργανον εἰμί epigram discussed above—is a missing link of sorts between a medieval syllabus such as the Schulgespräch and expressions in the ancient commentators that tread the line between bibliographical designations and references to topics or curricular functions of texts (such as Ammonius’ τὸ λογικὸν ὄργανον τῆς φιλοσοφίας discussed above).

This line was evidently finally crossed in Byzantine education. If for the ancient commentators the ‘instrument of philosophy’ was the topic of the ‘instrumental books’, it is no surprise that ‘instrument’ would end up being used as a title for those books. Perhaps this did not happen as early as the sixth century, but it also certainly did not happen as late as the fifteenth.