Every August, I, a hundred of my friends, and all the medieval re-enactment gear you could want spend 10 days at Raglan Castle, in Wales. It is an amazing castle, and a very special place with lots of memories as I’ve been doing this since 2008. The length of the event lends itself to leisure, and one of the things I love to do is bring along medieval texts to read aloud and discuss. This year, I brought along Asad Q. Ahmed’s translation of the section on logic of Avicenna’s Deliverance — a book which has been sitting on my shelf for almost four years but which I have not yet read. The Deliverance was written in 1027, a turning point in Avicenna’s logical career. Ahmed says in the introduction: “Somewhere around 1027, Avicenna starts to show less patience with Aristotle himself, frequently pointing out the failure to implement one set of principles consistently throughout the Organon.” The Deliverance, however, was a compilation of work written prior to 1027, and so reflects a more orthodox approach.
Slightly over half a dozen of us spent more than an hour on the text. Not having read any of it, I didn’t do any preparation in advance of the discussion, so some of our discussion centered around “Well, I’m not entirely sure what he’s referring to here, possibly X”. We read all of sections 1-8, and then I glossed over and summarized sections 9-40. Here are some of the interesting points that came out of the discussion.
In section 1, “On Conceptualization and Assenting and the Method of Each”, Avicenna claims that “all primary cognition and scientific knowledge is either conceptualization or assenting”. Conceptualization is acquired by definition, while assenting comes via syllogism. Both definition and syllogism are divided into (1) the real, (2) the unreal but “beneficial to some extent in its own way”, and (3) the “false that resembles the real”. He makes the interesting point that humans are not naturally disposed to being able to distinguish these three; but since logic involves being able to make these distinctions, that is why it’s important to study logic. As evidence for the claim that people, by their nature, are not able to distinguish these three types is that if this weren’t the case, then “there would occur neither any disagreement among the wise nor any contradiction in the judgement of any single one of them”.
Both definitions and syllogisms are hylomorphic, being composed of matter and form. What is interesting here is that in order for a definition or a syllogism to be a good one, both the matter and the form must be good in combination with each other. For “just as corruption in the building of a house may occur on account of the matter even if the form is correct or on account of the form even if the matter is sound (or on both their accounts together)”, so may a definition or syllogism have be defective either because the form is good but the matter is not; the matter is good but the form is not; or because neither is. This is a distinctive view to me (if anyone knows of anyone else who holds such a view, please share in the comments!), because usually the importance of the form of a valid argument is stressed because if the form is good, then the argument will be good regardless of the matter that is inserted. Unfortunately, Avicenna doesn’t give an example of a syllogism or definition where the form is good but the matter is not.
Having introduced the subject matter of logic, in section 2, Avicenna explains what the benefits of logic are. (Chapters like this are always my favorite, because they justify how I spend my life.) He first explains the different types of good definition and syllogism. There are two types of good definitions, true definitions and descriptions (which are merely convincing rather than true), while there are three types of good syllogism. The first is the correct type, and is called demonstration. The second is a convincing syllogism which “generates a kind of assent that resembles certainty”, and these are the dialectical ones. Then there is a weak type which “generates overwhelming belief”, and these are the rhetorical ones. Finally, there is the false definition, which is called misleading, and the false syllogism, which is called sophistical. A sophistical syllogism “presents itself as a demonstrative or dialectical syllogism, while not being so”. Finally, there is a fifth type of syllogism which does not generate any type of assent but rather effects the imagination, and this is called the poetic syllogism. (Side note: I’ve never heard of the poetic syllogism before. Is this in Aristotle? Who else discusses these?)
Avicenna then notes that the relation of logic to deliberation is the same as grammar to speech and prosody to poetry, but while “a sound nature and innate faculty of discernment can perhaps dispense with the study of grammar and prosody”, there is no substitute for the study of logic.
Sections 3 and 4 cover Simple Utterances and Complex Utterances. The former are those utterances which are significative on their own, no part of which is significative, while the latter are those utterances which are significative but which have significative parts.
Sections 5 and 6 cover Universal and Particular Simple Utterances, which are distinguished on the basis of whether they signify “the many by way of one coinciding meaning” or not. Here, “the many” can either be many in existence, such as “man” which signifies many men, or in the imagination, such as “sun” which signifies one existing sun but nothing prevents it from signifying in the imagination many other suns. Particular simple utterances are those “whose unique meaning cannot possibly be anything more than a unique thing”. Examples of particular simple utterances include proper names and deictic descriptions such as “this sun” or “this man”.
This is pretty much all Avicenna has to say about simple utterances at this point, and the next section, 7, is a long one dedicated to what counts as essential, because every universal utterance is either essential or accidental. An essential utterance “sets down the quiddity of that of which it is said”, which, you have to admit, is not an entirely helpful definition. A further gloss is provided that
the essential is such that, if the meaning [of the subject] is understood and occurs in the mind and if the meaning of what is essential to it is understood and occurs in the mind at the same time, it would be impossible for the essence of the subject to be understood unless first the meaning [of that which is essential to it] is already understood to belong to it.
From this it is clear that the essential is not that which is merely inseparable from its substance. For it is inseparable of a triangle that the sum of its angles equal two right angles, but this is not an essential property of a triangle, because one can understand “triangle” without necessarily first understanding “sum of its angles equally two right angles”. This can be contrasted with the example of “man” and “animal”; “animal” is essential to “man” because you cannot understand “man” without understanding “animal”.
Given all this, section 8 is a short discussion of the Accidental, which is that is not essential. After this point, we stopped reading aloud entire sections and rather I skimmed and summarized.
Sections 9 and 10 are on answers to the questions “What is it?” and “Which thing is it?” We skipped these sections and went on ahead to the classification of the five types of universal utterance (section 11), genus, specific difference, species, property, and accident, each of which is given their own section (12, 14, 13, 15, 16, respectively). Sections 18-20 cover nouns, verbs, and particles, and then we have definitions of statements (21), which are merely complex utterances; propositions (22), which are statements in which “there is a relationship between two things such that the judgment ‘true’ or ‘false’ follows from it”; and attributive propositions (23), where the two things being related can both be picked out by simple utterances.
Statements themselves can also be related to each other, and the result is a conditional proposition (24). There are two types of conditional propositions, conjunctive and disjunctive. A conjunctive conditional proposition is one like “if the sun rises, the morning exists” (25), while a disjunctive conditional proposition is one like “Either this number is even or this number is odd” (26).
After this we are introduced to concepts of affirmation and negation, subject and predicate, singular and indefinite propositions, and then into everything that goes into the Square of Oppositions. At this point we stopped reading from Avicenna and I grabbed a stick and began drawing in the sand. It was not very easy to photograph, but here you go: