Leibniz on Material Things
In this paper I describe some conflicting threads in Leibniz's arguments about the ontological status of material things such as blocks of marble. As is well known, in Leibniz's mature metaphysics the things real in the fullest sense, or absolutely real, are the monads, soullike simple substances which have no parts and which are self-sufficient in the sense that the internal states of a monad follow from each other without any external influences (monads are "windowless"). So in some sense, the reality is basically non-material, or mental. Material things of our everyday experience are somehow ontologically dependent, or supervenient upon the basic ontological level of monads. What is not so clear is how exactly the supervenient level (or maybe levels) of reality depend upon the basic level. What kind of a story can we tell about the way material things can be reduced ("revoco") into what they really are (AG 181)? Leibniz seems to have two stories to offer, one which starts from the perceptions of monads, and another in which the concept of aggregate plays the main part. Whether these two stories are really in conflict has been one of the eagerly discussed subjects in Leibniz scholarship (i.e., Adams 1994, Hoffman 1996, Rutherford 1995, ch. 8). On the following pages I try to tell the stories so that the incompatible ideas in them become visible, and after that I bring in enough of Leibniz's metaphysics so that we can see how the two stories can be related to each other. My thesis will be that the phenomenalist story is more basic of the two, the aggregate story requiring the other as a background.
Material Things as Phenomena
In a letter to Arnauld written in 1686 Leibniz says of things like a block of marble that they are "moral beings, beings in which there is something imaginary and dependent on the fabrication of the mind" (AG, 79). In another place in the same correspondence he critisizes "those who make realities out of all the abstractions of the mind". Only complete entities really exist, and blocks of marble do not belong to them; what God created when he created the world were the complete entities, "all the rest is merely phenomena, abstractions or relationships" (LA, 127). On the basis of these texts we might ascribe to Leibniz some sort of phenomenalistic view of matter: in trying to understand the ontological status of material things, we have to make reference to minds and their internal states, their perceptions. Perhaps the clearest formulation of these phenomenalistic tendencies in Leibniz's mature thinking occurs in a letter to de Volder:
If we do not go any further we get a picture of reality not too far from what is usually ascribed Berkeley. The existence of material things depends upon the fact that there are minds which have right sorts of correlations between their perceptions. Material things are not the outside causes of our perceptions; their esse is their percipi.
But there is one element in what is said above that is definitely not Leibniz and this has to do with the relation between what is inside a perceiver, i.e. perceptions, and what is outside of it. For Berkeley our perceptions of material things do not represent anything outside of us, but Leibniz thinks that all the perceptions of every monad express the rest of reality. Those perceptions in which material things are present to us do not correspond in any straightforward way with the reality outside of us, but that does not mean that there is nothing they represent. Material things as phenomena in our minds are phenomena bene fundata: there is in reality something which they express. What do they express then? The only real things there is, of course, the individual substances:
"Our mind notices or conceives of certain genuine substances which have various modes; these modes embrace relationships with other substances, from which the mind takes the opportunity to link them together in thought and to enter into the account one name for all these things together, which makes for convenience in reasoning." (LA, 126)
But this passage might give the wrong impression that Leibniz thinks that we are able to understand what it is really like beyond our experience. This is not the case:
"We do not comprehend the nature of odors and savours, and yet we are persuaded, by a kind of faith which we owe to the evidence of the senses, that these perceptible qualities are founded upon the nature of things and are not illusions." (Theodicy, G VI 74, quoted from Mates, 203)
So our experience of the external material world is not just a long coherent dream. In our experience the reality to which we take part is present to us, but only confusedly so that we are not conscious of the exact nature of this reality. It is like the experience of hearing the sea murmuring when approaching the seashore: what we really hear are single waves breaking but when these negligible perceptions come together to make our experience we can no longer be conscious of the elements it in fact consists in. (AG, 65)
Material Things as Aggregates
The other story of the relation between matter and monads goes differently. First of all we notice, following Leibniz, that material things cannot be substances, because they are extended and anything extended is always in principle further divisible, divisibility being a sign that the thing in question does not belong to the fundamental level of being. Whatever is divisible is not an unum per se and its reality is derived from the reality of its parts. If the parts of a material thing are themselves material, they again have to borrow their reality from their parts, and so on ad infinitum. If matter is to be anything real we have to end up to basic elements which are no longer divisible but unities by themselves. In Monadology Leibniz says that "there must be simple substances, since there are composites; for the composite is nothing more than a collection, or aggregate, of simples" (Monadology, par.2; AG, 213). And because these simples have no parts, they cannot have extension either. These "metaphysical points" are what we find, if we analyze matter completely. But at this point Leibniz emphasizes that it would be a mistake to think of these final elements, the monads, as parts of the aggregates. The part-whole-relation requires for parts to be homogeneous with the whole, i.e. the ontological status of parts is the same as of the whole, but in the analysis of matter it is impossible to find the final ground for the existence of matter if we do not make a jump from one ontological category to another of altogether different character (AG, 103). Monads are said to be the "elements" of the aggregates. This of course leaves pretty much in the dark the precise character of the relation between the two levels. But the picture seems to be that from the multitudes of monads there emerges subsets of monads, and the monads in these subsets constitute the aggregates there is in the world. Every aggregate has monads in it, in the sense that there are some particular monads which go to make it up. This story of material things as aggregates seems to ascribe them an ontological status different from the one ascribed to them by the phenomenalist story. We might try to remove the conflict by saying that whereas in the phenomenalist story Leibniz is using the word 'body' to refer to representations in the monads, he is now using it to refer to what those representations are representations of. The problem with this is that Leibniz expressly says that aggregates are phenomena. How is this possible? How can aggregates which have substances as their elements be identified with things inside each of those substances?
Bringing the Two Stories Together
We can start to relate the two stories together by concentrating our attention to what it means for Leibniz to say that aggregates are not composed of monads but result from them (AG, 179). One way to try to understand this is to take into account Leibniz's ontology of relations. It sounds natural to say that for a set of monads to form an aggregate, the monads in question have to be related in some way.
Here is Leibniz on relations:
"I do not believe that you will admit an accident that is in two subjects at the same time. My judgment about relations is that paternity in David is one thing, sonship in Solomon another, but that the relation common to both is a merely mental thing whose basis is the modifications of the individuals." (L 609)
According to this view the basic ontology consists of simple substances and their inner states, but when we begin to consider the relations between the simple substances we get into the realm of entia rationis, mental things. This sounds strange to modern ears but perhaps the following example will help: Let us think of two sticks with equal length. If we try to think in a Leibnizian way, we have to say that what really exists are the sticks and their properties, and the relation of equality between them is something that emerges when someone compares them. The relation is not there, it is not as real as the other constituents of the situation. Leibniz thinks of all relations in the same way.
This conception of relations leads Leibniz to consider all complex entities as not quite real; if something is not intrinsically one, an unum per se, it's existence is in some way mental or based on convention (LA, 126). Though, for the sake of argument, we would admit that individual sheep are real, we should not admit that a flock of them on field is real as well. Real sheep form a new entity, a flock of sheep, only for some mind which considers the individuals together. Of course, Leibniz admits that there might be a foundation in the properties of the individuals for linking them together. But the linking itself has to be done by a mind. This idea is strange: why couldn't we say that the linking of the individuals is done by the relations between them? Well, for Leibniz there is no such things as relations without a mind.
So the jump from the level of monads to the level of aggregates of them is at the same time a jump from the level of real things to the level of things which are mind-dependent or phenomena. We have found a reason for Leibniz to call aggregates phenomena and so a way to bring the aggregate story closer to the phenomenalist story. On the whole, however, I think that it is not very helpful to start with the aggregate story and try to understand Leibniz the phenomenalist from that standpoint. But the doubt remains that we still haven't find out how an aggregate of monads, of real substances, can BE a phenomenon, a mental thing. One way to put the problem is the following: the existence conditions of phenomena seem not to involve the existence of other monads than the one who experiences those phenomena, so they cannot involve other monads as any kind of constituents. An aggregate of monads, on the other hand, seems to require the existence of the particular monads which constitute it.
Adams who thinks that it is possible for phenomena to have real things as constituents, has tried to sooth the doubt by interpreting Leibnizian phenomena as intentional objects, and he thinks that whereas phenomena as inner states of minds could only have psychological properties, as intentional objects they can have any properties that bodies could have and can have real objects as constituents (Adams 223). Discussing the problem with Hoffman who has criticized Adams's solution, Adams has given the following example of how real thing can be elements in intentional objects (quoted from Hoffman 115):
But isn't it just ontologically misleading to speak of the actual Napoleon as a character in a novel. If it were discovered by future historians that there never were Napoleon Bonaparte, that all the "facts" about him are only a big lie, would it follow that some constituent of Tolstoy's novel were missing? I think not. Something would have been discovered about the relation between Tolstoy's novel and the real world, but that would not change anything in the novel itself.
I think that in the end the only possibility to get rid of the conflict is to play down the importance of the aggregate conception of bodies. We can see how this happens, if we start with Leibniz's basic ontology of monads and their inner states, and try to see how it is possible from that basis to understand that there is a diminished sense in which we can speak of monads as connected to some particular bodies. In this construction the bodies are ontologically speaking only appearances, they are things whose existence is totally dependent on their being perceived. At the start we might think of monads as monitors on the screens of which there is a film going on (the analogy of monitors is from Glenn Hartz, but he uses it for different purposes, see Hartz, 542). The monitors themselves belong to a common world only in a very thin sense; we should try to think of them as separately as we can, corresponding to the separateness of Leibniz's monads for which there is no common space in which they could be thought to exist "together". Perhaps we could think the world of monitors as the set which has them as elements. The films on different screens are not the same, but there is some kind of resemblance between different films so that God who sees them all can understand them as depicting the same series of events but from different points of view. We should even think that God could take one film as his starting point and as it were calculate the films on the other screens, i.e. if God sees what the world represented looks like from one point of view, he would know what it has to look like from any other possible point of view. At this point there is no sense in which the thin world of the monitors and the world represented in different ways on their screens are connected. As far as we (or God) can tell, the world of the films might be only a dream, or pure fiction. Now there is one thing in Leibniz's theory about monads and their perceptions which is still missing from our analogy. In Nouvaux Essais Leibniz expresses it thus:
"Every finite spirit is always joined to an organic body, and represents other bodies to itself by their relation to its own body. Thus it is obviously related to space as bodies are." (NE II, xv, 11; quoted from Rutherford, 208)
The experience of monads is in other words essentially indexical. We shouldn't describe experiences by sentences like 'monad a sees a tree' but by sentences like 'monad a sees a tree over there' or 'monad a sees a red ball in front of itself'; that is objects present to us in our experience are always seen as related to one particular body which we experience as our own. And in this way we can say that monads as it were sink into the world of their experience. It is this sinking which creates a connection between the phenomenal world and the world "in" which the monads exist. It is not only that every monad feels itself to be part of its own phenomenal world: because of the harmony between the perceivers every monad gets a place in the common phenomenal world, or in the world which every monad sees from a particular point of view. In other words it becomes possible to say that monad a perceives monad b in the sense that there is a body x in the phenomenal world of a, which corresponds to the body y in b:s world which b perceives as its own body. And finally we get a sense in which monads can be said to constitute aggregates: a set of monads perceives as their own those bodies which in the common phenomenal world of the monads are spatially close together. So we can construct the aggregate story from the phenomenalist building blocks. What we have to remember here, however, is that the aggregate story describes the reality at the level somewhat above of the fundamental level.
AG: Ariew, R., and Garber, D. G.W. Leibniz: Philosophical Essays. Indianapolis: Hackett, 1989.
L: Loemker, L.E. Leibniz: Philosophical Papers and Letters.
Dordrecht: D. Reidel Publishing Company, 1976. LA: Mason, H.T. (trans.) and Parkinson, G.H.R. (intro). The Leibniz-Arnauld Correspondence. Manchester: Manchester University Press, 1967.
Adams, R.M. Leibniz: Determinist, Theist, Idealist. Oxford University Press, 1994.
Hartz, G. 'Leibniz's phenomenalisms', The Philosophical Review 101 (1992).
Hoffman, P. 'The Being of Leibnizian Phenomena', Studia Leibnitiana, Band XXVIII/1 (1996).
Mates, B. The Philosophy of Leibniz. Oxford University Press, 1986. Rutherford, D. Leibniz and the Rational Order of Nature. Cambridge University Press, 1995.