In the last article, we saw how William of Ockham developed his nominalist philosophy in the context of disputes within the medieval Franciscan order. Ockham’s nominalism—the thesis that there are no real, abstract universal concepts, but that these terms refer only to ideas that we have—undercut Aristotelian arguments about the naturalness of property ownership, based as they were on the assumption that abstract universals such as “human,” and therefore some such thing as “human nature,” really did exist. If there is no such thing as an essential human nature, neither can there be any such thing as certain, permanent, a priori knowledge, on the model of geometry and as advocated by Plato, about how humans ought to live, including the propriety of property ownership. So nominalism strengthened the hand of the Franciscans who wanted a more austere, poverty-embracing way of life, against the popes, bishops, and their allies in the Dominican order, who thought that the naturalness of owning property could be deduced from self-evident truths about human nature. If there was no human nature, those “truths” were not self-evident, but only names. There were many other facets to this discussion, but the general idea of nominalism was to attack these kinds of knowledge claims, those based on abstract, speculative, metaphysical reasoning, which had characterized Western philosophy since the time of Plato.
Nominalism seemed to cut human beings adrift, however. If universals exist, are evident to reason, and can be used as a basis for extracting certain, permanent knowledge, then there is a close connection between the intellect and reality. Getting at truth may be difficult in the sense that one has to put the work in, but it is not difficult in the sense that the concepts of truth, or of knowledge, are in themselves problematic. Some ancient philosophers had taught this, of course, but Plato’s supremacy ensured that it never became the dominant position. For most philosophically inclined people, they would not become problematic again until Hume and Kant raised the specter of skepticism in the eighteenth century. But it was rather the point of Platonic philosophy that certain knowledge was, finally, obtainable. One simply had to define one’s terms, and think carefully about what those definitions entailed.
There aren’t going to be any big surprises with a God who is predictable, rational, and more or less fully known.
But if all our concepts are mere names, as William of Ockham taught, then the kind of knowledge that Plato sought was not available. In a modern context, where reason is the only path to knowledge, this would have entailed some form of skepticism. If reason is all you have, and all reasoning proceeds on names and appearance, then the kind of knowledge that one can expect to acquire is going to be very limited. It might have all sorts of advantages, but it is not going to be “truth-with-a-capital-T,” the certain, final knowledge that Plato and his followers sought. So a certain degree of skepticism is built into a nominalist outlook. In a naturalistic context, and one accommodated to nominalism for so many centuries, this is simply par for the course. But in William of Ockham’s context, where Christian theology occupied a very high station, it could be a truly terrifying idea.
The reason is that Platonism made God intelligible. God was not arbitrary or capricious, according to a Platonic conception, but preeminently rational, and hence knowable. A God that can be known is, in a sense, a safe God. That kind of God produces regular effects in a regular manner, so one’s obligations are clearly spelled out, and one simply has to follow the rules in order to stay in God’s good graces. There aren’t going to be any big surprises with a God who is predictable, rational, and more or less fully known. So, by emphasizing the rationality of God, Platonic philosophy emphasized the reasonability of God, and hence of reality itself. It was a reality where one could feel at home.
But if all concepts are in us, not out there, then what do those concepts mean when applied to God? Their meaning becomes uncertain, so God’s activity and character, too, become uncertain. If, for instance, someone says that God is just or loving or wrathful, then we know what that means by way of analogy to humans. Such a God acts the way humans do when they are just, loving, or wrathful, so there aren’t going to be any big surprises. But if those words do not mean the same thing when applied to God that they do when applied to human beings—after all, the concepts are in us, and don’t necessarily refer us to anything real in the world, let alone in God—then what do they mean? Well, they could mean anything. Perhaps God’s love is very different than human love. Perhaps God’s justice and wrath are very different as well. So there was a danger in the nominalist worldview, that concepts would become detached and uncertain in a way that, on the Platonist worldview, they were not.
Where Thomas Aquinas, following Aristotle, emphasized God’s rationality and omniscience, William of Ockham emphasized God’s freedom and omnipotence. “God is no man’s debtor,” Ockham frequently said, meaning that He is not obligated to human beings in any way. God cannot be so-obligated, according to Ockham, because that would constrain His omnipotence, which is a logical impossibility. So God can create a world one day, and uncreate it the next; God can make promises one day, and unmake them the next; God can save this one, and damn that other one, for any reason, or for no reason at all. Thus human beings are never in a position to question God, in Ockham’s view.
The concepts of omnipotence and inscrutability, taken to their logical extremes, meant that God could never quite be trusted.
This conception of God could be a very unsettling prospect. It was made much more plausible during the outbreaks of plague that ravaged Europe continually during the late medieval period. By some estimates, up to a third of Europe’s population was wiped out, and the medical technology of the time was more or less powerless to prevent, or even understand, these deaths. What possible meaning could such events have? It really seemed that God might, out of sheer caprice, destroy anyone at any time for any reason. Under the influence of nominalism, God became less a universal spirit of love, as experienced by the first Christians, but a continual threat to human well-being. The concepts of omnipotence and inscrutability, taken to their logical extremes, meant that God could never quite be trusted. Only a very radical kind of spirituality (the kind that, for instance, might induce someone to join the Franciscan order, and renounce all one’s worldly property) could feel any devotion toward this kind of God.
Ockham’s conception of God had some important consequences for science, however. If the Platonic conception of God was true, then one could reason from God to the world, and from the world to God, because everything happened out of logical necessity. In fact the Roman Catholic Church never entirely embraced this view, since a God who not only could not contradict logic (as most Christian theologians have held), but was positively forced, at all times, to act from logical necessity, would not be free in the sense required to affect any kind of reconciliation or harmony with the Hebraic, prophetic tradition, as contained within the Bible. The Hebraic conception of God is of an emphatically free being, who is in relationship with the chosen people. But if every act of God is logically determined, then how is God free? So the Roman Catholic Church never embraced the conception of God as a being whose every act was logically determined, as Spinoza was to do in the seventeenth century. But it remained the case that one could reason a good deal about God from the state of the world, and the state of the world from what one knew about God. Especially in the Aristotelian, Scholastic philosophy of Thomas Aquinas, the rationality and omniscience of God was definitely privileged over the freedom and omnipotence of God. So a priori knowledge about the world is obtainable, on that view.
Nominalism involves an important shift in the concept of knowledge itself. A priori knowledge doesn’t count anymore
But Ockham’s nominalism called this assumption into question. If God is essentially a volitional, rather than a rational, being, and if God’s omnipotence could have produced a world of virtually any conceivable description, then one could not just sit back and reason about how things must be. One has to go out and look. True, such knowledge will only be partial and provisional, but since that’s the only kind of knowledge that is available in any case, according to nominalism, one can hardly complain about it. So nominalism involves an important shift in the concept of knowledge itself. A priori knowledge doesn’t count anymore.
To give an example of the difference in thinking involved, Bertrand Russell once mocked Aristotle, saying that he “maintained that women have fewer teeth than men; although he was twice married, it never occurred to him to verify this statement by examining his wives’ mouths.” How can something so obvious have escaped the attention of the great philosopher? Well, on the Platonic conception of knowledge, just counting the number of teeth in his wives' mouths would not have gotten to the heart of the matter. All it could have told him is how many teeth were in the mouth of a particular woman. He could have counted all the teeth in all the mouths of the people in his village, and the result still wouldn’t have been knowledge. It would have told him about the teeth in the mouths of particular people, not about human mouths as such. Because, for Plato, the world of sense experience, that where things come into being and go out of being, is not the real world, but only a shadowy reflection of the forms that have true existence, the kind of empirical inquiry that Russell called for could not, in principle, establish knowledge. Real knowledge is about real objects—which is to say, the forms. And it has to be certain, and permanent in character, not mere conjecture. Real knowledge is the kind that we have about triangles: they always and necessarily have three sides, and interior angles of one hundred and eighty degrees, not any other number, ever, else it is not a triangle. So if geometry sets the standard for what counts as knowledge, Bertrand Russell’s suggestion that one simply count teeth is not applicable. One first has to have the concept of nominalism in place to think that this kind of procedure is going to produce knowledge. In other words, one has to already be accustomed to the idea that knowledge is only ever partial and provisional, to think that counting particular teeth in particular mouths is going to yield results that ought to count as knowledge. Aristotle didn’t look into his wives' mouths and count their teeth, not because he was ignorant, but because he was not a nominalist.
By shifting the grounds of knowledge, Ockham’s nominalism helped pave the way for modern science: by privileging empiricism (that is, observation and experiment) over rationalism (that is, an emphasis on a priori deduction from first principles); and by encouraging us to think of the provisional and partial nature of the resulting knowledge, not as disqualifying it from the category of genuine knowledge, as Plato held, but rather, as both a humbling and an exhilarating aspect of our pursuit of it. It is humbling, because one has to be willing to think of the knowledge that results as always open to future revision, so one can never be quite certain that one has laid a hold of truth. And it is exhilarating, because it means that the voyage of discovery is, in principle, without end.
This essay is part of a series; the previous essay can be found here.
Daniel Halverson is a graduate student studying the History of Science and Technology. He is also a regular contributor to the PEL Facebook page.
I am really learning a lot from these essays. Thank you for putting them together. I would think that nominalism works against science. Science, as I understand it, is an attempt to understand the universe through empirical, objective categories. However nominalism denies the existence of any objective categories. Therefore wouldn’t nominalism lead to anti-realism, and as a result undermine science?
Thanks for your kind words, Michael. I’m glad you’re enjoying the series.
This is an interesting philosophical point that you raise. I’ve given some thought to it, and this is how it seems to me. In the 19th century there was a little-remembered philosopher of history named Wilhelm Windelband. He argued that one had to distinguish the nomothetic from the idiographic sciences. Nomothetic sciences are what we usually think of when we think of science. They’re law(nomos)-seeking, and they’re characterized by a general invariance across time and space. Physics, for instance, works the same on earth as in the Virgo supercluster, will be the same in ten billion years as it was ten billion years in the past. A physical law, in order to count as a law, has to be independent of any particular object, yet applicable to all of them. Chemistry is also a nomothetic science.
Idiographic sciences are more numerous, but we usually don’t think of them as having a basically different explanatory structure. Following Windelband, I would argue that they do, however. Nomothetic sciences describe particular, temporally- and spatially-coordinate objects. The descriptions they yield are not necessarily, and do not have to be, valid across time and space. They just have to be accurate with respect to the particular object under consideration. In geology, for instance, what we want to know about is the planet earth. The question whether the theory of plate tectonics is applicable to planets always and everywhere doesn’t arise, because it’s not the science of planets just as such, but of the planet earth in particular. In paleontology, we want to know the history of life on earth specifically. We don’t concern ourselves with life as a property of the universe. Idiographic sciences are historical in the sense that they deal with objects as they exist in time and space, rather than in the abstract.
You see the connection to realism (that is, about abstract objects or universal categories) and nominalism. It’s difficult to see how we can have a nomothetic science without thinking that there really are such things as laws of nature. Yet thinking in terms of laws does not always provide the best results. There is no law that I’m aware of that says that the earth has to produce so much a quantity of igneous rock and so much a quantity of basalt, or that a meteor had to hit the earth and wipe out the dinosaurs thus leading to the rise of mammals. Basically, the more individualized and complicated our objects of study, the more nominalism and the idiographic conception of science it implies is going to help us. The simpler the objects (quarks, for instance) the more realism is going to help us.
In modern philosophy there’s a strong emphasis on monism. There’s a sense that we have to try and reduce things to the smallest number of principles possible – preferably, one. It seems to me that a strict monism creates a lot of problems, however. I think we can be realists about some things (mathematics and physics, for instance) and nominalists about others (planets, animals, human beings.) Taken to an extreme, nominalism does undermine the nomothetic sciences in the sense that it denies that there really are physical laws, and also in the sense that it seem to imply the fictitious character of the mathematics on which those physical laws rest. A very strong nominalism implies an anti-realist skepticism about the nomothetic sciences that, in my view, doesn’t really do them justice. But a hard swing to realism creates its own problems. It forces us to think in terms of abstract essences, such as “species” and “human nature,” where the evidence seems to be for contingency and particularity rather than for universality. (To clarify, about species, Darwin denies the real existence of species as categories – according to Darwin, species are just breeding populations, each of which is continually in flux. We just can’t see it because the change happens on a very large timescale, relative to an individual human life.)
So I think a monistic approach, asserting the exclusive truth of either realism or nominalism, creates more problems than it’s really worth. I think ontological dualism provides the most satisfactory answer here. But if we grant some room for both, but nevertheless insist on making a choice in terms of priority – which ought to be our default position, so to speak, unless faced with compelling reason to adopt the other – then I think nominalism is more often than not the applicable theory. In articles on the controversy over Darwinian biology, where exactly this issue of universality vs. particularity, realism vs. particularity, comes up, I’ll try to show how nominalism has proved the more fruitful approach, at least in evolutionary biology. But you’re right, a really strict nominalism seems to undermine confidence in science to discover truth.
Daniel
Hi Daniel,
Thanks for your response. Even if idiographic sciences deal with more local, particular phenomena, from how you describe it there still does seem to be an acknowledgment of objective categories. For example in your description of idiographic sciences, there is still talk about objective, mind independent categories such as “objects” and “planets.” So it seems there is still the use of metaphysical categories and natural kind terms. Would’t true nominalism deny the existence of even these?
Great thread; and agreed with Michael’s assessment of the value of this series. Particularly the serial nature, built in substantive but not overwhelming blog-bites, is helpful for sustained engagement and comprehension.
And my guess, Michael – though we’ll have to see how Daniel replies – is that he will agree with your rejoinder as a limit case, but also argue that – even in nominalism – we must build these mind-independent categories… if for no other reason than to successfully explain the point of nominals. If I am surmising correctly, he will assert that those are somewhat artificial / constructed entities themselves, basically meant for our cognitive aid.. and playing to our basic human disposition to favor categories over so many unique things.
My additional speculation: Perhaps Platonism was so favored for so long because it’s intuitively satisfying to our organization / making sense of a pretty complicated world.
So, Daniel, just to explore the nominalist conception of women’s teeth a bit more. Well, we’ve gone out and counted the number of teeth in the mouths of many, many women. And it appears to us that, save a very small margin of variability, almost all women have the exact same amount of teeth as men are typically observed to possess.
In this case, we are not going to call that near certainly of finding X teeth a universal abstract – and therefore a form – .but rather just an observed regularity in what are otherwise nominal entities (i.e. discrete things) we’ve provisionally-labeled as human beings. Is that the right way to look at this puzzle, from a faithful, nominalist perspective?
Yes, that would be a nominalist way to look at things. From a nominalist point of view, this counts as knowledge because it’s based on observation and may be provisionally relied upon for practical purposes. From a realist point of view, it wouldn’t, because real knowledge is universal (that is, without qualification with respect to time and space), and no amount of observation can, even in principle, establish that this is the case.
Daniel
Oh, and I’m glad you’re enjoying the series. More to come!
Hi Michael,
As you point out, nominalism involves an ironic posture toward language, the use of concepts in which one does not actually believe because it is convenient or necessary to do so. There is a sense that our concepts never quite latch onto the way the world is, and a certain degree of skepticism is unavoidable as a result. These are, arguably, drawbacks to nominalism. The payoff is flexibility in linguistic usage (if concepts don’t really exist outside of us, we are free to invent or discard them as our needs require) and the elimination of interminable disputes over definitions (since the concepts are in us, not in the world, their definitions become matters of convention.)
The reason I say that nominalism makes more intuitive sense when we come to idiographic sciences is that they are concerned with change in a way that nomothetic sciences are not. The term “object” implies a kind of stasis or solidity which, when we consider phenomena as extensions not only across space, but also across time, does not seem very descriptive. In this case, an “object” is perhaps better described as an “event” – an ephemeral and dynamic concatenation of causes. But this would be an overly-technical and awkward way to talk in everyday life. One doesn’t normally refer to stars and candy canes and poodles as “events,” although, from a historical point of view, that does seem to be what they are.
There are plenty of people who think nominalism is false, and who don’t subscribe to an idiograhic/nomothetic distinction, so I don’t want to imply somehow that these would be unphilosophical positions. Good and persuasive arguments can be made for and against.
Daniel
So what would some of these drawbacks (to taking a nominalist view of the world) be, Dan? That we are therefore impoverished – both philosophically and practically – by never being able to make universal claims about anything? That nominalism puts on the inevitable road to relativism? That trying to talk intelligently about regularities / universals out in the world merely reduces us to a language game? Or, even, that building a systematic view of the world becomes impossible when ideas and things become unmoored from fixed meanings?
That’s it, in a nutshell.
Right; thanks Dan. So what is the sharpest rejoinder to these objections? Can nominalists fall back on some principle of statistical significance, perhaps, to defeat the straw-man argument of relativism / nihilism? i.e. That while we still can’t speak to universals from this worldview, there is still enough observable regularity out there that we have discretion / license to make real claims, if those still within some notable margin of (inevitable) error… which we just chalk up to the imperfect nature of the world?
Sure, that’s one way to go about it. There are a lot of potential solutions to this challenging and important question in philosophy. Kant thought we could make a division between the phenomena and the noumena. Perhaps a probabilistic/Bayesian approach is the best. Maybe we should just bite the bullet on anti-realist skepticism. Maybe proximate knowledge is all we can really have. Then again, maybe nominalism is false.
In my view nominalism leads to some good insights but needs to be moderated in order to avoid extreme and probably false conclusions. In general, philosophy presents us with a choice between consistency and comprehensiveness. A really consistent philosophy will probably leave some important things out. Materialism, for instance, doesn’t seem to have a good account of mind. Nominalism doesn’t seem to have a good account of mathematics. etc. But if we want comprehensiveness, it’s hard to stay consistent. We can be realists about mathematics and nominalists about other things, but since realism and nominalism are competing ontologies we aren’t answering the question about the most basic type of being in a consistent way. Philosophy is in a sense a series of choices about which intuitions we think are the most important and ought to be preserved.
Daniel
And maybe also why the tradition goes on and on and on, with ambitious systematizers always being subsequently overturned. This riddle of choice (between consistency and comprehensiveness) seems like an interesting philosophical dilemma unto itself.
Kant himself, if I am not mistaken, was in some sense trying to reconcile a traditional realist / Platonic conception of the world with nominalism, no? To marry Reason with observed, scientific method, which was starting to displace the default deductive approach with inductive methodology?
Yes, that’s right. Kant’s division of being into the phenomenon and the noumenon was an attempt to accommodate the very different approaches to knowledge implied by realism and nominalism under one synthetic system. Daniel
Good stuff; thanks again for your blog-posts, Daniel! Having this exchange with you confirms that I am learning something. Appreciate it.
Daniel,
I’ve been studying medieval philosophy for a series of articles I’m writing, and a friend of mine pointed me to your blog, here.
It may be due to my naivety and lack of philosophical training, but to me it seems that nominalism is self-contradictory.
If it is, I would have bet the rent that my objection to it is already well-known and perhaps refuted.
But on the off-chance that I am onto something, here is my “proof”…
Given: Boethius’ definition of a universal.
“A universal has to be common to several particulars in its entirety, and not only in part, simultaneously, and not in a temporal succession, and it should constitute the substance of its particulars.” (From Stanford Encyclopedia of Philosophy)
Here is why nominalism is absurd:
1. Philosophy consists largely of truth claims.
2. All truth claims presuppose the existence of truth.
3. One such truth claim is called Nominalism: that it is true there are no universals, but only particulars.
4. Truth, if it exists, is a universal (it meets the definition).
5. But nominalism claims that universals, and therefore truth, do not exist.
6. This implies that, if nominalism is true, then nominalism cannot be true.
7., Therefore, nominalism is absurd.
And thus, by the rules of nominalism, each universal would have to be examined on a case-by-case basis in order to deny each one. That would keep Ockham busy for a while.
And at the end of the day, it would imply that all thoughts and truths are subjective, including this one. And if God exists, even His thoughts.
Humans have objective knowledge and subjective thoughts. I think we need a different term for God’s level of omniscience. God does not have opinions, has no need for them. He has truth. That’s why Jesus said, “I am the truth.” It’s a philosophical statement. The truth exists and is embodied in the Persons of the Trinity. Acquiring knowledge consists of learning what God already knows — getting closer to His perspective. His “subjective” thoughts are objective truth from our perspective.
In other words, universals exist if God sees it that way, And the Bible seems pretty clear about His thinking just that way.