The Problem of Universals
The problem of universals was first clearly recognized and dealt with by Plato. Here is one passage where he brings the problem out clearly (Euthyphro 5c9-e2, 6d9-e6). Philosophically speaking, this passage is a mess. By emphasizing different parts, this passage could be interpreted as affirming any of the four main philosophical positions taken in answer to the problem of universals. But before we look at answers, let's see what the problem is.
At 5c, Socrates asks Euthyphro what piety is, and Euthyphro's reply is say that what he is doing is piety. He was prosecuting his own father for killing one of his day-laborers. Euthyphro extends his answer to include other things relevantly similar. Piety is not only what Euthyphro is doing right now, but piety is any action like his, e.g. if someone else prosecuted a murderer, that would be piety, if someone (Euthyphro or anyone else) prosecuted a temple thief, that would be piety. In answer to the question, "What is piety?" Euthyphro gives a list of actions.
At 6d, Socrates points out a problem with the answer. Socrates didn't ask for a list of pious actions, he asked for an eidos or an idea. Before we consider what those words could mean, begin by paying attention to the language. When Socrates asks Euthyphro at 5c what piety is, he uses a noun: "piety." When he corrects Euthyphro at 6d, he uses an adjective: "pious." Euthyphro failed to distinguish the noun from the adjective. Socrates didn't want a list of things that could be described with the predicate adjective "pious," he wanted something that could be stated using the predicate nominative "piety." Verbally we make this distinction in English all the time; e.g. "happy" is the adjective, "happiness" is the abstract noun; "kind" is the adjective, "kindness" is the abstract noun; "courageous" is the adjective, "courage" is the abstract noun. The Greek language draws the same distinction.
I've drawn attention to this because this gets right to the core of "The Problem of Universals." The problem is whether this verbal distinction is purely verbal, or whether there is something deeper to it, whether this verbal distinction corresponds to anything in reality. Is this distinction a real distinction? Is there a real difference between "being pious" and "being piety," between "being courageous" and "being courage," is there a real difference between "being feline" and "being felinity." Are abstract nouns just words and nothing more, or is there something real which they name? The former view is called "Nominalism" and the latter view is called "Realism." So now let me clarify the difference between Realism and Nominalism.
Before I define these positions, let me pursue the general description of the problem of universals a bit further. Look back at Socrates' question to Euthyphro in 5cd. He doesn't just ask about piety, he also asks about impiety. Ultimately (6e) he wants an answer that will let him sort things into two groups: put all the pious actions over here in one pile, and put all the impious actions over there in a different pile. If there are actions left over, they will make up a third pile: actions that are neither pious nor impious. Words that group things into piles of different sorts are sometimes called "sortal predicates." Concepts according to which things are grouped together are sometimes called "sortal concepts." The things grouped together are called "particulars" because they are the "particular" things in the group. They are also sometimes called "tokens." The grouping is sometimes called the "type." So for example, suppose you have 19 coins in your pocket, and you sort them into different piles: you might have 10 pennies, 3 nickels, 4 dimes and 2 quarters. In that case, you have 10 tokens of the type "penny," and so on. You could also say that you have 10 particulars of the type "penny." You could also say that you have 10 instantiations of the universal "penny."
Now we are almost to the heart of "the problem of universals." Think again about grouping things together. For example, think about grouping all the coins in your pocket. You could group them by denomination, as I just described. But you could also group them according to the metal they are made out of. In that case, you would have 10 copper coins, and 9 silver coins. You could also group them according to the year in which they were minted. There are a great many different ways you could group the 19 coins in your pocket. But let me ask you this: is there a right and a wrong way to group them? Is there a way of grouping them together in piles that is objectively correct in such a way that if you grouped them together differently you would be grouping them together incorrectly, or unnaturally? If you are inclined to answer this question "no," then you may be inclined towards "Nominalism." If you are inclined to answer this question "yes," then you may be inclined towards "Realism."
Let me ask the same question in a slightly different way. What is clear is that you have 19 items. There are 19 particulars, 19 individuals, 19 tokens. The particular individuals clearly exist. What is not so clear is whether the type, the sortal, the universal "coin" exists in addition. In a pile of 19 coins, do we actually have 20 entities: the 19 particulars plus the one universal "coin"? Does the universal "coin" actually exist somehow? If so, what denomination is it? Where is it located? Is it legal tender? Now here are the definitions of "Nominalism" and "Realism."
Nominalism =df everything
that exists is a particular individual, no universals exist.
Realism =df in addition to particular
individuals, some universals exist.
The word "Nominalism" comes from the Latin word nomen which means "name." According to Nominalism, universals are nothing more than names, or words we use to group things together. According to Realism, at least some universals are more than just names we use to group things together, some of the names for universals actually name really existing entities.
Plato was clearly a Realist about universals. His most famous metaphor for the reality of universals was to say that real universals "cut nature at its joints" (Phaedrus 265d-266a). He compares the task of definition to the job of being a butcher. The clumsy butcher just hacks things up in any old way, but the expert butcher deftly slices the animal at its natural joints, neatly separating naturally distinct segments of the animal. He gave an extremely influential example of this method of definition in Sophist 218d-221c.
This is the original inspiration for what is now called "taxonomy." The Greek word "taxis" refers to the arrangement of a group of soldiers. Biological taxonomy is the study of the arrangement of living organisms into the categories: Kingdom, Phylum, Class, Order, Family, Genus, Species. Plato's idea here is that the various species are real features of the world. Tobermory my cat really is a cat (Felis catus). By distinguishing cats from dogs we "carve nature at its joints" and are recognizing a real distinction which exists in nature. This is the realist view of species.
Now that I've covered the basic idea of the problem of universals, and have defined the two main positions, let's go back to the Euthyphro passage and see what indications there are as to how Socrates wants to solve the problem of universals.
So start out, Socrates asks Euthyphro, "What do you say is pious and impious in murder and other cases?" The focus is on how Euthyphro talks. Socrates wants to know which actions Euthyphro labels with the word "pious," and which he labels with the word "impious." This suggests the position called Predicate Nominalism.
Predicate Nominalism =df everything that exists is a particular individual, universals are merely general words applied to more than one individual.
You might call this "pure" or "strict" nominalism. According to this view, only particulars exist, and the only sort of reality that universals have is in the fact that we tend to make the same sound when confronted with different things. I breathe through my voice and make the sound "desk" when I see that individual there, but I make the same sound when I see that individual over there. That's all there is to universals. The only philosopher ever to explicitly espouse this view was Roscelin of Compiègne (1050-1125). He is famous for saying that a universal is nothing more than a flatus vocis, a "breathing through the voice." His two main opponents were St. Anselm and Peter Abelard.
We don't know much about Roscelin's arguments for his view, but probably it was similar to the central reason given later by John Locke in favor of nominalism in his Enquiry Concerning Human Understanding iii.3.
1. The greatest part of words are general terms. All things that exist being particulars, it may perhaps be thought reasonable that words, which ought to be conformed to things, should be so too,- I mean in their signification: but yet we find quite the contrary. The far greatest part of words that make all languages are general terms: which has not been the effect of neglect or chance, but of reason and necessity.
2. That every particular thing should have a name for itself is impossible. First, It is impossible that every particular thing should have a distinct peculiar name. For, the signification and use of words depending on that connexion which the mind makes between its ideas and the sounds it uses as signs of them, it is necessary, in the application of names to things, that the mind should have distinct ideas of the things, and retain also the particular name that belongs to every one, with its peculiar appropriation to that idea. But it is beyond the power of human capacity to frame and retain distinct ideas of all the particular things we meet with: every bird and beast men saw; every tree and plant that affected the senses, could not find a place in the most capacious understanding. If it be looked on as an instance of a prodigious memory, that some generals have been able to call every soldier in their army by his proper name, we may easily find a reason why men have never attempted to give names to each sheep in their flock, or crow that flies over their heads; much less to call every leaf of plants, or grain of sand that came in their way, by a peculiar name.
We use general words rather than coming up with proper nouns for every particular thing. Instead of addressing this snowflake with the name "Jasper," and that snowflake with the name "Sally," we just lump them all together with a single word an call them all "snowflakes." On this view, there is no reality to this grouping.
Probably the main reason that no philosopher other than Roscelin ever adopted this view is that it seems to make all of our groupings entirely arbitrary. We could just as well have a single general term that applied to this desk, that wall, Toby my cat, the moon and Dean Martin. But that's absurd. Plato was clearly onto something in his quest for general definitions because not all of our groupings are entirely random. Surely it is not completely arbitrary that we put Toby in the group of cats and not in the group of tea kettles, or the group of planets. Strict Predicate Nominalism hasn't been popular at all.
Back to the Euthyphro for an alternative. Socrates doesn't stop at what Euthyphro calls "pious" or "impious." At 6e, he asks Euthyphro for a paradeigma that he can use to distinguish the pious actions from the impious actions. If he holds up the paradigm to this action, and it matches, then the action is pious, if it doesn't match, then it is not pious. The idea seems to be that a universal is an exemplar or a standard, a model or a pattern. For example, suppose you work in the produce section of a grocery store, and the produce manager wants you to set up a display of all the best apples to entice people to buy apples. The manager tells you one to pick the apples that are really red. She may select one in particular and say, "Like this." You then take your exemplar and match other apples to it; if their color resembles the color of the exemplar, then you pick it to use in your display; if it doesn't match, you don't use it.
This suggests a different kind of nominalism. On this view, there is more to a universal than just a name. Although the only things that really exist are particular individuals, some of those individuals resemble each other.
Resemblance Nominalism =df everything that exists is a particular individual, universals are merely resemblances between individuals.
This is perhaps the most popular Nominalist alternative to Predicate Nominalism. One of the most important theorist who seems to have accepted this is John Locke. Here is a brief statement part of his view in his Enquiry Concerning Human Understanding iii.3.
7. Shown by the way we enlarge our complex ideas from infancy. But, to deduce this a little more distinctly, it will not perhaps be amiss to trace our notions and names from their beginning, and observe by what degrees we proceed, and by what steps we enlarge our ideas from our first infancy. There is nothing more evident, than that the ideas of the persons children converse with (to instance in them alone) are, like the persons themselves, only particular. The ideas of the nurse and the mother are well framed in their minds; and, like pictures of them there, represent only those individuals. The names they first gave to them are confined to these individuals; and the names of nurse and mamma, the child uses, determine themselves to those persons. Afterwards, when time and a larger acquaintance have made them observe that there are a great many other things in the world, that in some common agreements of shape, and several other qualities, resemble their father and mother, and those persons they have been used to, they frame an idea, which they find those many particulars do partake in; and to that they give, with others, the name man, for example. And thus they come to have a general name, and a general idea. Wherein they make nothing new; but only leave out of the complex idea they had of Peter and James, Mary and Jane, that which is peculiar to each, and retain only what is common to them all.
Locke makes a big deal of this point because he thinks it is one of the central flaws of Medieval philosophy that it spent so much time debating the reality of universals. Several times in the Enquiry Locke criticizes the "Schoolmen" or the "scholastics" for their endless wrangling over this problem. One of the most famous (or infamous) was Duns Scotus, known as the "Subtle Doctor" (Doctor Subtilis). His defenders were known as The Dunsmen, or, later "Dunses" or "Dunces." This was a term used to criticize them as people who refused to keep up with the new learning, but insisted on sticking to the outmoded philosophy of the Middle Ages.
Resemblance Nominalism is clearly more defensible than strict Predicate Nominalism, but it raises an important question: how is resemblance to be judged? In my story of the produce manager, the paradigm was a particular physical object. We might call this "Paradigmatism."
Paradigmatism =df everything that exists is a particular individual, universals are merely resemblances between individuals to one particular paradigmatic individual.
You might try to defend this view by an appeal to developmental psychology. If you want to understand the nature universals, or from the Nominalist's point of view, general terms, more fully, you should look to the origin of general terms in the psychological development of children. When children learn sortal terms, e.g. "dog" and "cat," they learn them by "ostention," i.e. by having examples pointed out to them. I probably learned the reference of "cat" when my mother pointed to my childhood cat Simba and said, "Look Donnie, that is a cat. Can you say 'cat'?" I learned to generalize this word by learning to identify things that were similar to Simba my cat.
Perhaps this makes developmental sense, but it suffers from a serious problem. If we learn the meanings of general terms by ostention, then we all learn different meanings for general terms. For me, the general term "cat" means Simba, and anything relevantly similar to Simba. For you, the general term "cat" means Fluffy, and anything relevantly similar to Fluffy. But since your childhood cat is not my childhood cat, then when you use the word "cat" you mean something different than what I mean when I use the word "cat."
Of course the reply is to say, "But Fluffy and Simba are similar to each other, they resemble each other, and so the meanings of your "cat" and my "cat" resemble each other. That is a reasonable reply, but it is a reply that a Paradigmatist cannot make. If the two resemble each other, then there must be some third cat whom they resemble. But since you and I have never seen this third cat, it cannot be what our words "cat" mean. So the paradigmatist seems stuck with the conclusion that you and I mean different things by the word "cat."
But another view is suggested by Plato in the Euthyphro passage. He uses the Greek word idea, which means "form" or "shape." This is where we derive the English word "idea." This is in fact what Locke chooses as the focal point for his resemblance theory (Enquiry iii.1).
3. To make them general signs. But neither was this sufficient to make words so useful as they ought to be. It is not enough for the perfection of language, that sounds can be made signs of ideas, unless those signs can be so made use of as to comprehend several particular things: for the multiplication of words would have perplexed their use, had every particular thing need of a distinct name to be signified by. To remedy this inconvenience, language had yet a further improvement in the use of general terms, whereby one word was made to mark a multitude of particular existences: which advantageous use of sounds was obtained only by the difference of the ideas they were made signs of: those names becoming general, which are made to stand for general ideas, and those remaining particular, where the ideas they are used for are particular.
According to Locke, the resemblance is not to some particular external object; rather the resemblance is to some internal, mental idea or conception. We might call this "Concept Nominalism," or "Conceptualism."
Conceptualism =df everything that exists is a particular individual, universals are merely resemblances between individuals and general concepts.
Instead of trying to match things to some particular paradigmatic exemplar, the way we group things together is by matching them to concepts we have in our minds. This avoids the problem with paradigmatism. You and I don't have to have learned the word "cat" by being confronted with one and the same cat. It is enough for you and I to have the same concept of a cat.
Of course this raises a different problem. How could I ever really know what your concept of a cat is? At least with paradigmatism, if you wonder what I mean by "cat" I can just pick Simba up and show you, or at least show you a picture of Simba. But I can't do the same with my concepts. I have to describe my concept to you. But if I use other words to describe my concept to you, how do I know that you have the same concepts of those other words as I do. Empiricists have a long history, well into this century, of offering various solutions to this problem.
But there is a very different way to interpret Plato's Greek words idea and eidos. These "forms" might be perfect paradigms that do not exist in our souls, they have an "extra-mental" existence, but they are not particulars: they may be really existing universals. This is the view that is universally associated with Plato, universals are "Platonic Forms."
Transcendent Realism =df in addition to particular individuals, some universals exist, and they have an existence separate from the individuals that instantiate them.
Pious actions are pious by "participating" in the form of piety. Sticks that are of equal length are equal by "participating" in the form of equality.
Unfortunately, transcendently real universals raise many deep problems. Plato himself brings out some of them in his later dialogue Parmenides. Central to these problems is this idea of "participation." How exactly do particulars "participate" in universals? Are bits of the universals chopped off and put into the particulars somehow? Is the whole universal somehow present in each particular? Another difficult puzzle is to explain just how the form of equal could have an existence separate from particulars. Consider biological universals, e.g. "human being." What could it mean to say that the universal "human" exists separately from particular human beings like you and I. What could it mean to say that even before human beings existed, the form "human being" existed?
But the Euthyphro passage doesn't require this extreme interpretation. Go back to the root senses of the Greek words idea and eidos as shape. Think of making gingerbread cookies with cookie cutters: every cookie made with the same cookie cutter has the same shape, so that shape exists. When all those cookies are eaten, that shape no longer exists, except in the cookie-cutter itself. That may be how universals exist. The universal "humanity" exists because there are particular individuals who are "human-shaped." In this case, since humans are not made with human cookie-cutters, if at some point the human race is entirely wiped out, then the universal "human" would cease to exist. This position is associated with Aristotle.
Immanent Realism =df in addition to particular individuals, some universals exist, but they do not exist separately from the individuals that instantiate them.
To summarize this, there are two main kinds of Realism and two main kinds of Nominalism. The first kind is Platonic Realism, often called "Transcendent Realism" or "Ante rem Realism." According to this view, universals are not only real, they "transcend" particulars in the sense that they have an independent existence from them. "Ante rem" means "before the thing." The second kind of Realism is Aristotelian Realism, often called "Immanent Realism" or "In rebus Realism." According to this view, universals are real, but they are not transcendent, they cannot exist independently of particulars. There are two main kinds of Nominalism. First, there is strict Nominalism or pure Nominalism, or "Predicate Nominalism," according to which only particulars exist, universals are merely a flatus vocis, or just a general name applied to distinct individuals. Alternatively, another form of Nominalism is called "Resemblance Nominalism," according to which the groupings we make of particulars are not real, but are based on something other than the way we talk about the world. We group things together because they resemble each other more or less. There are two main variants of Resemblance Nominalism. First, there is Paradigmatism, according to which the resemblance of particulars is judged by some paradigmatic or exemplary individual. Second, there is Conceptualism, according to which the resemblance of particulars is judged by some general mental concept.
There is perhaps no other topic than the Medieval problem of universals that is more maligned by both philosophers and non-philosophers. In a passage where he seems to have in mind Dun Scotus (known as the "Subtle Doctor," and in ¶8 Locke mentions the "subtilty" of Scholastics in the title of the paragraph), Locke criticizes the Scholastics for their useless debates (Enquiry iii.10).
9. This learning very little benefits society. For, notwithstanding these learned disputants, these all-knowing doctors, it was to the unscholastic statesman that the governments of the world owed their peace, defence, and liberties; and from the illiterate and contemned mechanic (a name of disgrace) that they received the improvements of useful arts. Nevertheless, this artificial ignorance, and learned gibberish, prevailed mightily in these last ages, by the interest and artifice of those who found no easier way to that pitch of authority and dominion they have attained, than by amusing the men of business, and ignorant, with hard words, or employing the ingenious and idle in intricate disputes about unintelligible terms, and holding them perpetually entangled in that endless labyrinth. Besides, there is no such way to gain admittance, or give defence to strange and absurd doctrines, as to guard them round about with legions of obscure, doubtful, and undefined words. Which yet make these retreats more like the dens of robbers, or holes of foxes, than the fortresses of fair warriors: which, if it be hard to get them out of, it is not for the strength that is in them, but the briars and thorns, and the obscurity of the thickets they are beset with. For untruth being unacceptable to the mind of man, there is no other defence left for absurdity but obscurity.
This is a common objection to Medieval philosophy in general, but the problem of universals lies at its heart. Often people criticize the Medievals for debating the question of how many angels could dance on the head of a pin. The common retort is, "Who cares!?" (Note the exclamation point.) Renaissance and Early Modern philosophers created a serious bias against Medieval philosophy, and so I really should say something about the importance of the problem of universals.
First and foremost, the problem of universals has theological importance. The position you take on universals can seriously affect what you can say about doctrines like the Trinity, transubstantiation and the existence of angels. For example, if everything that exists is particular, and unity of kind is not real unity, then it is hard to see how God could be one being while simultaneously being three individual persons. The only real unity would be the three, and not the one. This would be the heresy of Tri-theism, and will get you excommunicated from the Church.
Realism and nominalism seriously affects our understanding of angels. For example, if you are a realist, and you think that two things of the same form are differentiated by their matter (you and I are both identical in that we are humans, but we are distinct humans because your matter is separate from my matter), then how many angels are there? Since angels are not material beings, matter cannot differentiate them, and so there can only be one angel! That is not orthodox, and it can get you into trouble.
Finally, the doctrine of transubstantiation is hard to deal with if you are a nominalist. Usually the doctrine is explained by saying that the accidental qualities of bread and wine persist, while their substances are removed and replaced with the substances of Christ's body and blood. It is sort of like the trick of pulling out the table cloth without disturbing the dishes on top, but it is more like pulling out the entire table and substituting a different one. This distinction of substance and accident can be made sense of in realist terms, but it is not clear how a nominalist could deal with them in an orthodox manner.
But beyond Christian theology, the problem of universals is extremely important historically for the development of scientific methodology in particular, and epistemology in general. A very important Renaissance scientist and philosopher, Francis Bacon, wrote a very important work which he titled Novum Organum(1620). Remember that Aristotle's logical works were called the Organum. Bacon wanted to free himself and science from the outdated methods of the Aristotelian past. He thought that science needed a new mode of inductive reasoning.
Induction which proceeds by simple enumeration is a childish thing, and concludes precariously, and is exposed to destruction by contradictory instances, and makes is pronouncements on the basis of too few instances, and then only from those which are present. But induction which will be useful for discovery and demonstration in the sciences and the arts ought to separate nature by rejections and exclusions and then, after a sufficient number of negatives, to conclude affirmatives. This has not been done so far, except to some extent by Plato, who certainly used this form of induction to investigate definitions and ideas. Here is a metaphor he used to describe this "new tool."
Empirics are like ants who gather and use; Dogmatics are like spiders who spin webs out of themselves. The bee is the true middle way: it gathers material from garden and field flowers and then with the faculties proper to her, she turns and digests.
The Scholastics seemed to Bacon to be like spiders, spinning his theories out of his head with little regard to actual data. If universals are real, then we don't have to spend much careful attention studying particular individuals: they are just the means we use to understand the universal forms they instantiate. This process is much too quick, according to Bacon; "to the human intellect we must add, not feathers but more lead and weight." Scientists needed to slow down, and pay much more careful attention to details. Instead of just lumping all beings into 10 Aristotelian Categories, scientists ought to go out in the world and look to see how many types of things there are. Look carefully at the details, instead of just blindly lumping things together into your pre-determined categories.
Clearly there is much to be said in favor of this view. It is also true that very soon after this period, science began to make some very important, and very rapid progress. Perhaps it was Aristotelian Scholasticism that had halted development for so on.
But there is a price to be paid for this new methodology. About a hundred years after Bacon, Hume (1711-1776) pointed out the problem.
The bread, which I formerly eat, nourished me; that is, a body of such sensible qualities was, at that time, endued with such secret powers: But does it follow, that other bread must also nourish me at another time, and that like sensible qualities must always be attended with like secret powers? The consequence seems nowise necessary.
If we want to be very careful and not lump things into the same category, if types are not real, if the only real things are particular individuals, then there are no general truths about bread. We can describe the color, shape, texture, taste and so on of this piece of bread, but if the general kind "bread" isn't real, then whatever I learn about this piece of bread won't help me learn anything about the next piece of bread. That is the crucial usefulness of real types: if "cat" is a real type, and not simply a nominal type, then whatever I learn about this particular cat will help me understand all cats. I can learn and know something about how to cure a problem with your cat if I have studied other cats, as long as they are identical in nature. If there is no reality to their unity as cats, then every new particular is just a new thing, and we can learn about it only by studying it; nothing else we study can possibly help us. So the existence of universals turns out to have a very profound impact on scientific methodology and epistemology.
Finally, I will just briefly mention something that all too often is either overlooked, or disparaged. Many philosophers are just plain curious about the structure of reality. Even if there are no practical consequences, even if it won't bring about peace in the middle east, or produce a cure for cancer, the problem of universals poses interesting questions about the fundamental nature of the universe in which we live. If you are just plain curious, it is one of the most fascinating of all philosophical problems.