(a) Everything that exists is one - a single thing.
This Augustinian (but not neo-Platonic, note) metaphysical principle makes, or at least appears to make it impossible for there to be anything common in the way universals are supposed to be common. Recall how Boethius said they were supposed to be common, in his Commentary on Porphyry: as a whole, simultaneously, and in such a way that they enter into the metaphysical structure of the things to which they are said to be common.
(b) On the other hand, all knowledge - or at least all knowledge anyone cares about - proceeds in terms of general concepts.
The problem of universals comes down to a choice. You can either choose to opt for the objectivity of knowledge, and say that there are general things corresponding to general concepts. The problem then is that you appear to make the world metaphysically impossible for the sake of preserving our knowledge of it. This is what realists do. Or: You can opt for a coherent metaphysics in which everything is singular, but you do so at the expense of knowledge. This is what nominalists do. So the question is, where do you want your problem? In metaphysics or in epistemology? You pay your money and you take your choice. Or - and this is what usually happens - you try to strike some middle ground and avoid the two extremes. (This may of course just turn out to mean that you have your problems in both metaphysics and epistemology.)
These are the historical dynamics of the situation. It goes without saying that I am not committing myself to saying that either the realists' metaphysical problems or the nominalists' epistemological ones are ultimately insoluble. But their solutions, if any, are not obvious - except to their partisans, of course. And in any event, this is, historically, where the problems lie.
Now realism was a view some people actually held. (We will see in a bit that there are some doubts whether anyone was ever a thoroughgoing nominalist in the twelfth century - or in the entire Middle Ages, for that matter.) There were certain theological considerations that led some people to think that an orthodox Christian had to be a realist. One such consideration was the doctrine of original sin.
Look at the passage from Odo of Tournai's treatise On Original Sin. First a word about the man. Odo of Tournai is also known as Odo of Cambrai. Odo died in 1113. The name 'Odo' is the equivalent of 'Otto'.
Look at what he says: The problem is this: We get our bodies from Adam, by biology. That is, there is a physical continuity between Adam and us - genetic continuity, although of course Odo wasn't thinking in terms of chromosomes. But the soul does not come from Adam. We do not inherit the soul. It is not transmitted in this way. On the contrary, the theological view was that the soul is subject to a kind of special on-the-spot creation by God at some time during the development of the fetus (not necessarily at conception). Now since sin inheres in the soul and not in the body (although the consequences of sin, such as disease, may inhere in the body), how can I be said to have inherited original sin from Adam?
Notice, incidentally, that this is a very good question. As a result, some people have toyed with the idea that we do inherit the soul from Adam. The doctrine of original sin is one of the most utterly fundamental doctrines of the faith, but it is a real mind-bender.
How does Odo try to explain the situation? Well, he says that it is possible to have a species that happens to contain only one individual in it. For example, take the sun. There could have been many suns. (Don't think of the word 'sun' as a proper name for the one we've got. And don't think of the sun as just another star. Think of it in the medieval way, as an especially large and bright body in our sky. There might have been several such bodies.) If there were many suns, they would all have the same substance or species, and would differ only by their accidents (line 25).(This recalls the doctrine of individuation by accidents in Boethius' De trinitate.) But in fact of course there is only one sun. (Odo's example is actually put in terms of the legend of the phoenix. But the principle is the same, and the sun example is the one other authors usually give to illustrate the point.)
Notice that Odo argues in lines 25-27 of the passage that you cannot in the same way have a genus with only one species in it. The whole idea of a species is that it subdivides a genus into a subsection. But it is not at all clear that Odo has a point here. I think he is probably confused. It is true that conceptually, if you add "rational" to "animal" to get "man", then you are implicitly allowing that there is a species of irrational animals too. Otherwise, the addition is not really an addition. But that's all on the conceptual level. It still seems possible to have it in fact be the case that the only animals that exist are, say, human beings - just as, as Odo himself admits, it was the case once (in the Garden of Eden) that the only human being who existed was Adam. In short, genera and species are alike on this point. If we are talking conceptually, then just as you can't have a genus that is not divisible into species, so too you cannot have a species that isn't divisible (potentially common) to several individuals. On the other hand, on the real level, just as you can have in fact a species with only a single individual in it, so too it seems that you could have a genus that turned out to have only a single species in it (although conceptually, of course, it could have more.)
Be that as it may. In any case, in lines 43-46 of the passage, Odo says "And when the species is said of a solitary individual, only then is it valid to attribute an accident both to the individual and to the species, although principally and in the first place accidents are in individuals."
That is, (1) first, accidents do not strictly speaking inhere in species. And where do you suppose he got that claim. Remember my claim that every theory of universals ever held in the Middle Ages can be found, in some non-trivial sense, at least in germ in Boethius? Well, we are beginning to see why I said that.
(2) Nevertheless, even though strictly speaking accidents do not inhere in species but only in individuals (Boethius would say "in the matter"), still, if there is only one individual in the species as it happens, and if it has an accident A, then we may in a sense (but not "strictly", of course) say that the species has A - because every individual in the species has it. What Odo is doing here is saying that if it happens that every B is an A, then we can in a sense say that the species or universal B itself is an A. If Adam had blue eyes, then as long as he was alone (that is, until Eve came along - and if she had blue eyes too, then the point would still continue to hold), it was in a sense true to say that humanity had blue eyes - since every human did. It is clear that Odo wants to generalize this, so that no matter how many individuals you have in the species, if all of them have an accident A, then we can in a sense say that the species itself has A.
Now here is how all this gets applied to the case at hand. (See the paragraph beginning in line 49 of the passage.) While Adam and Eve were still the only people in the garden, they both sinned. That is, every human sinned, and therefore, by (2) above, humanity itself in a sense sinned. This sin in a sense infects the whole species, since no part of the species - no individual - did not sin. Hence we, who are of the same species as Adam and get souls of the same nature as his (although of course we do not get them biologically from him), end up with souls infected with sin. By the time a third human being appears on the scene, it is already too late. The entire nature has been affected.
It is important of course that both people sinned. If Eve had eaten the apple and then given it to Adam who threw it away in pious indignation, then there would have been no original sin in the whole species. Presumably Even would have been in big trouble, but she alone would have been guilty. In order to affect the entire human species, Adam's contribution was crucial too.
But of course this is a crazy doctrine. If you poke at it even a little bit, it falls apart. If Adam and Eve were both right-handed, then so is the entire human species. If Adam and Eve both danced a jig in the garden of Eden, then so do I, since what they did affects the whole human nature, in which I partake.
But my reason for discussing Odo's theory is not that it is a good theory, but rather that it is a strongly realist theory. And notice, please, why Odo thinks he needs to be such a realist. It is to account for the doctrine of original sin.
This was a theological reason for realism. There are philosophical reasons too. Perhaps the most basic philosophical reason comes from a view about the correspondence of our thoughts and judgments to reality. That view is as follows:
If our knowledge is to be objectively grounded in the world, then there must it seems be a one-for-one correspondence between the elements of our judgments and things in the world. (This is sometimes called the "picture-theory" of language, or of truth. Our true judgments somehow "mirror" the world.) What grounds the truth of our judgments in the world must have exactly the same structure as our judgment itself has.
Those who don't like views like this sum it up by saying that, on such a theory, the world turns out to look very much like an English sentence (or German, or whatever other language the authors of such theories happen to speak). And in a sense this summation of the view is not too far off the track, although it is not so much the structure of the spoken languages that is at stake, but the structure of judgments. (This by itself suffices to disarm a number of objections.) The most basic motivation behind realism - at least the most basic philosophical motivation - is the view that the form of our true judgments is the logical form of the world.
Why does realism have to lean on some such principle as this? Well, remember, realists historically have always been mainly concerned with preserving our knowledge of the world. Now if you grant that the relation between our knowledge and the world does not have to proceed in this one-for-one fashion, that it can be more indirect, then why do we have to have a common or universal entity to correspond to our general concept? In short, why do we need realism?
I do not deny of course that a subtle theoretician might reject the extreme form the "correspondence" principle I have just given you, and yet still find philosophical reasons for thinking he has to be a realist. Nevertheless, the principle I have stated encapsulates a prominent underlying historical tendency in its most extreme - and therefore its clearest - form.
The extreme form of this principle is not just a historical abstraction; I am not just setting up a straw men. Some people in the Middle Ages did in fact explicitly hold a relatively extreme form of the principle. For instance, a certain Fridugisus, in a very odd Letter on Nothing and Shadows.
Note that there are several ways to spell the man's name. Concettina Genmnaro, Fridugiso di Tours, (Padua: Editrice Dott. Antonio Milani, 1963), pp. 67-96, discusses no fewer than seventeen variant spellings, each with some manuscript authority. They are (p. 71): Fredegis, Fredegisis, Fredegissus, Fredegisus, Fredegysus, Fredugisus, Fridarius, Fridegisus, Fridigisus, Fridogisus,, Friducis, Fridugerius, Fridugilsus, Fridugisus (by far the most common spelling in the manuscripts, and the one I adopt), Fridugusus, Frigdugisus, Frudigisus. Such variation in the spelling of proper names is not at all uncommon in the Middle Ages.
Fridugisus is from the ninth century, and was associated with the court of Charlemagne, which puts him shortly before Eriugena. Nevertheless, it is all right to treat him out of order, because the point I want to make about him only comes up now. Fridugisus was not an important philosopher.
In lines 27-28 of the translation: "Every finite name signifies something." The word 'name' here is a medieval grammatical term meaning both nouns and adjectives. A "finite" name is contrasted with an "infinite" name, which is a name like 'non-man', 'non-tree'. (See the discussion in Aristotle, De interpretatione, 2, 16a30-32, which is where this terminology comes from.) Fridugisus doesn't say anything about whether infinite names signify something; his claim is confined to the finite ones.
Now 'nothing' is a finite name, as the grammarians tell us. It is not an infinite name, since it does not have the prefix 'non- ' or an equivalent. (In Latin, 'nihil' = 'nothing' is not obviously made up of 'no' plus 'thing', as in English. Etymologically, to be sure, the word comes from 'ne' + 'hilum' = 'not in the least', but let's just pretend we don't know that.) Hence it must be a finite name. After all, it functions grammatically as a noun. It can serve in subject position, for example. Hence, according to the claim "Every finite name signifies something", 'nothing' signifies something. But just as to say that 'tree' (note the single quotation marks) signifies a certain kind of plant, or to say that 'dog' (again note the quotation marks) signifies a certain kind of animal amounts to saying that a tree (no quotation marks) is a certain kind of plant and a dog (again no quotation marks) is a certain kind of animal, so too, it would seem, to say that 'nothing' signifies something entails that nothing is something. And that certainly seems peculiar. (Quotation marks did not appear until the invention of printing. Medieval authors had other ways of marking what we do by quotation marks.)
But Fridugisus just accepts the consequence, peculiar or not, on the basis of his principle about finite names. In fact, he goes on to say that this nothing is a very important something, since it is that out of which God created every- thing. God created ex nihilo, remember.
You see what is happening here. The extreme form of the "correspondence" principle I set out above gets you involved in hopeless confusions and paradox. Recall Augustine's trouble over 'nothing' in The Teacher (Hyman and Walsh, p. 22). Also, Anselm's worries over nothing (if I may put it that way) in Monologion, Ch. 8. Eriugena's five types of nothing are perhaps based on a similar kind of principle. Eriugena's interpretation of the claim that creation is ex nihilo.
You might object that of course this is all silliness. We sophisticates in the twentieth century all know that 'nothing' may look like a noun, and grammatically is one (actually, it's a pronoun, but the point is the same) - it can be the subject of a sentence, for instance. But of course really it can be paraphrased away. To say that God created ex nihilo is only to say that he created and did not create from something.
That's true. But now you're tampering with the "correspondence" principle. You are saying that there need not be a one-for-one correspondence between true thought and the world - or rather the parts of true thoughts and the parts of the world. The correspondence can be more indirect, got at by paraphrase, for instance. But by weakening the principle in this way, you may well be weakening the pressure toward realism.
There is of course another problem with this extreme form of the "correspondence" principle, this time a theological problem. If every finite name signifies something in this straightforward way, then privative terms like 'blindness' signify real entities, not just the absences of other real entities (to wit, of sight). And worse, 'sin' and 'evil' are names and so signify real entities. Hence, we are back to Manicheanism.
In this connection, it is curious to note that Odo, who is a realist on other grounds, nevertheless rejects Manicheanism and says that evil is no real thing. (See PL 160, cols. 1072C-1073D, in Book I of his On Original Sin. He is not being exactly inconsistent here. You don't have to be a realist on only philosophical grounds. Odo is a realist on theological grounds (original sin), but tempers his realism on the basis of other theological considerations (he wants to avoid Manicheanism).
Now let's look at this again. I repeat, I am dealing here with what might be called the "pressures" of the situation rather than with strict philosophical necessities. Historically, realism has been philosophically motivated by concerns to save our knowledge of the world. You are going to think this is a problem that only realism can solve only if you also think there must be a "correspondence" between our true judgments and the world. And the simplest and most straightforward way to view that correspondence is in terms of the principle I enunciated earlier. If you take that principle seriously, at face value, then you are going to end up with nothings and evils in the world. There are theological pressures, therefore, against taking the principle too strictly. How all this works out in the case of a particular thinker is of course entirely dependent on his own wits.
It is of course perfectly possible to be fairly sophisticated about it, and to say that the surface structure of our judgments masks a deeper and more fundamental structure, so that words like 'evil' and 'nothing' can be strictly done without in a suitably pure language. (Recall the "ideal"-languages of some early-twentieth century philosophers.) And then you might say that the correspondence between thought and reality does not occur at the level of surface structure, but at the level of deep structure. Then you might go on to say that, while at that deeper level of fully expanded paraphrase, you may not need words like 'nothing' and 'evil' - and so do not need real nothings and evils as positive beings in the world - you certainly will need some general terms at any rate, and so some universals in the world.
It is possible to say all this, but it is fancier than things got in the twelfth century. It will become more explicit later in the Middle Ages. William of Ockham (early-fourteenth century), for instance, will argue by means of such considerations of "deep structure" or paraphrase (these come out in his so called "connotation theory"), that you don't need entities corresponding to the words 'nothing' and 'evil'. He went even further and said that you don't need entities to correspond to terms in all the Aristotelian categories. You can say everything you want to say - although no doubt not so briefly - by using only substance terms and quality terms (and connecting words). You don't need real places, for instance, to answer to terms in the Aristotelian category of place. And so on. Motion for Ockham is not something over and above substances and their qualities. You don't need such motion terms in a suitably paraphrastic language, and so you don't need such entities in your ontology.
Ockham uses the paraphrase argument in this way to cut down the list of categories in the world. But when he comes to argue that, even in the categories of substance and quality, where you do need general terms to say everything you want to say, you nevertheless don't need real universal things in the world, he has to turn to a different kind of argument entirely. Here he has to reject the "correspondence" principle even in its sophisticated, "deep structure" form.
So much now for realism. While realism was held in the twelfth century in various more or less strong forms, I am not so sure that nominalism was ever held in a pure form in the twelfth century - or at any other time in the Middle Ages. But it has been attributed to a certain Roscelin, by Anselm and by others.
We have some letters to and from Anselm concerning Roscelin. The points in all these places are mainly theological, but it is clear that Roscelin is supposed to have held a fairly strong form of the view that there is nothing shared or common in reality - not even in God. Hence if there are three divine persons, then there must be three gods, and there is nothing at all that unites them. Roscelin is the only medieval figure I know of ever to have been accused of out and out tritheism. (See, for instance, the Letter on the Incarnation of the Word, lines 64-65.) It is doubtful that he ever really went that far. (It also doubtful that the view attributed to him is coherent. Questions of realism and nominalism in connection with the Trinity are extremely obscure and difficult.)
John of Salisbury is a little more helpful. At the end of the twelfth century, John tells us (Hyman and Walsh, p. 167) that Roscelin thought that "universals" were mere "word sounds". The Latin here is 'voces', plural of 'vox'. That is, there are general terms, general in the sense that they are predicable of many things, but there are no general or universal realities corresponding to them - that is, general in the way a metaphysical universal is supposed to be general. (Recall Boethius' characterization of how this would be.)
The term 'vox' is important here. In the Letter on the Incarnation of the Word, line 36, Roscelin is said to have held that "universals" were mere flatus vocis ('flatus' is a fourth-declension plural), rather delicately translated there as "verbal puffs". I'm sorry, people, but yes, it can also be translated "verbal farts" - I didn't make up the term. And, given the context, and given that folks in the Middle Ages were not overly refined about that sort of thing, I'm not at all sure that this isn't exactly the sense Abelard had in mind.
A vox ("'utterance' is probably a good translation) is simply any sound uttered by the vocal mechanism of an animal - and therefore excludes the stamping of feet and the breaking of trees, as some people rather colorfully put it - and that can be written down, that is, spelled, thus ruling out sneezes and coughs and so on. (This last is what they meant by saying that a vox is an "articulate" sound.)
Now a vox need not be significant, it need not mean anything. An occasional medieval example later on was "Bu ba blitrix", which doesn't mean any more in Latin than it does in English. John of Salisbury contrasts Roscelin's doctrine with Abelard's, who says that a universal is a sermo - rather hard to translate very well, but it means a significant or meaningful vox. That is, Abelard wants to say that while there are general terms, there are no universal realities corresponding to them, and nevertheless they are
significant. They are grounded in the world, so that knowledge is saved. We will look at this more carefully later on.
In any case, if John of Salisbury is right in his sketch of Roscelin, then Roscelin didn't care whether a vox is significant or not. (Otherwise the contrast with Abelard loses its point.) What Roscelin leaves out of his account is how thought or language is significant, how it links up with the world. As far as Roscelin's doctrine goes - or at least as far as John's description of it goes - everything in the world is singular, and if that compromises our knowledge, so much the worse for knowledge.
I list these two items simply because they are good, and you should know about them.
Secretum secretorum index - De Mirabilibus Mundi index
Atlas Games - publishers of Ars Magica Redcap - Ars Magica portal
Last modified: Sun Dec 27, 1998 / Jeremiah Genest