Home | Philosophy

Spinoza and Substance Monism
By: IrvingWashington

“In the universe there cannot be two or more substances of the same nature or attribute.” (The Ethics, 1p5)
Discuss.

1p5 is a proposition vitally important to Spinoza’s metaphysics. In this essay I am going to discuss his attempted demonstrations in 1p5dem and 1p8sch2. I contend that both demonstrations fail. However, I do think the proposition makes sense within Spinoza’s system, and I will attempt to indicate why. To begin, it will be necessary to explain the meaning of three key terms, and to place 1p5 within the context of Book 1 of The Ethics.

The three key terms are substance, attribute and mode. What precisely Spinoza means by these terms is a controversial matter. However, for our purposes this isn’t important. I propose we understand these terms in what would be the natural way, i.e. similar to the use in Descartes and other philosophers of the period. So: a substance is both ontologically and epistemically independent of anything else. In contrast a mode is both ontologically and epistemically dependent on a substance. All this means is that a mode requires a substance in which it exists and it requires the prior conception of a substance before it can be conceived. Every substance has attribute(s) . Put simply, an attribute is the nature of a substance: the type of substance it is. To make this clearer, a (un-Spinozistic) example: our body is a substance with the attribute of extension (by its nature an extended thing). Its modes are just the determinate way in which it exists, its shape, size etc – what makes this body a body of a certain type. We cannot conceive of the shape of the body without the body itself, i.e. the substance, and the shape can’t exist without the body. In contrast, the body requires nothing else for its existence, and can be conceived through itself. There are complications with attribute I pass over, for my purposes an attribute expresses an essence or nature of something.

1p5 plays a major role in Spinoza’s argument for substance monism: there is only one substance, God (or Nature). The first fourteen propositions of The Ethics are intended to establish substance monism, and the argument can be summarised as follows :
(1) 1p5: There cannot be two substances of the same attribute.
(2) There exists a substance with infinite (i.e. all) attributes.
(3) Consequently, there can be only one substance, as no substance can have no attributes, any other supposed substance would need to share one of God’s attributes, so contradicting (1).

The argument for substance monism isn’t what concerns us here; I shall focus on 1p5. However, it should be clear what is at stake for Spinoza here. He needs 1p5 to establish his substance monism, without it his official argument fails . I shall now examine his argument for 1p5, starting with 1p5dem.

1p4 states that two substances are distinguished either by a difference of attributes or of affections (modes). This is uncontroversial: there exists nothing other than substances (with their attributes) and modes. The difference must lie in one or the other. There are two parts to the argument for 1p5, corresponding to the two ways we can distinguish. The first part deals with a difference of attributes, which I shall label the argument from attributes. Basically, all Spinoza says is that if two substances are distinguished by a difference of attributes then we obviously have two different substances, as they have different attributes (remember, an attribute expresses what a substance is). There is a problem with this, attributed to Leibniz . Say we have two substances, X and Y, where X has attributes {A1, A2}, Y has {A1, A3}. Clearly X and Y have something in common (A1) but are distinguished by virtue of possessing a different second attribute. Generally, the problem is that for cases with substances of more than one attribute Spinoza’s argument just doesn’t work. I will briefly return to this later.

The second part deals with a difference of affections. The argument, which I shall label the argument from affections, is less clear, stating:

“if they are [two substances] distinguished by a difference of affections, then, since substance is by nature prior to its affections (Pr.1), disregarding therefore its affections and considering substance in itself, that is (Def.3 and Ax.6), considering it truly, it cannot be conceived as distinguishable from another substance.”

Obviously, if we can disregard the affections the conclusion follows. The question is why can we do this? The key inference is clearly from ‘considering substance in itself’ to ‘considering it truly’. As Don Garrett points out , this inference is not valid. By ‘in itself’ here is meant we can conceive substance as not existing in something else (as a mode does). To conceive something truly is to have an idea of it agreeing with the object of the idea (1Ax6). The claim is that substance conceived in itself, in the above sense, is substance considered truly. This is false: substance doesn’t exist ‘in itself’ in the sense that its not with anything else, a substance will always have certain affections i.e. ways that it is. The substance is truly with its affections, so a true idea of a substance must include them. I am going to focus on this problem, for reasons that shall become clear.

In the literature there are many attempted solutions. Don Garrett, in his Ethics 1p5, gives a nice summary, and reasons for rejecting. What interests me is his solution, which is the best attempt I have found. I shall present it, then show it fails.

Garrett begins by asking what the priority of substance over its attribute involves. As stated earlier, it involves a mode being both in and conceived through its substance. We could disregard the affections if the relation of mode to substance was such that any difference in affections would be conceived through a difference of attributes (for our purposes here we can take substance and attribute as synonymous). The basic idea is that if the affections of substance X differ from the affections of substance Y without this difference being conceivable through a difference in the attributes of X and Y themselves then there will be something about the affections of X not conceivable through X. It will be helpful to state this in argument form.

(1) The affections of a substance X are completely conceived through X (1def5).
(2) These affections cannot be completely conceived through X unless all of their features are conceived through X.
(3) Features of affections cannot be fully known (or conceived) without knowledge of their cause.
(4) The knowledge of the cause of its effect is knowledge of what determines the effect to happen.
(5) Any feature in which the affections of X differ from Y must be conceived through some difference in the attributes accounting for this difference.

This argument seems valid. Consequently, the distinction in affections is only an apparent one. Since we only had a distinction of affections, we don’t have a distinction of attributes. Thus, part of 1p5 is established. Garrett himself helpfully points out the assumptions involved here, and answers an objection he feels one might have. Rather than consider all of this, I wish to consider one assumption that I think we should reject. (1) and (2) can be seen as leading to what Garrett calls a ‘strong’ version of 1def5, which is that any feature of the affections of X can be conceived through X. The feature we are here talking about is that the affections of X differ from those of Y. But, if we wish to include this feature it seems that we run into problems. How can this be a feature of the affections of X? That the affections of X differ from those of Y is surely a feature of X and Y jointly, not X alone. At this point nothing stops us positing another substance, Z. Say the affections of Z differ from those of X: this is now, for the same reason, a feature of X. And so on ad infinitum. Does X possess the feature for every such substance we can posit? Also, how are we meant to conceive of a difference of affections through a difference of attributes? Do we conceive of the affections of Y through X, and so compare? But Spinoza’s system does not allow this. Maybe we conceive of the affections of X through X, of Y through Y, but in what sense are we conceiving of the difference of affections through a difference of attributes here? But my main point is that a difference of affections between X and Y is not a feature of just X, or of just Y, and so the argument does not go through. This objection to me seems valid. So, Garrett’s attempt to rescue this part of the argument for 1p5 fails. As I mentioned, none of the other attempts look convincing so I will conclude that no attempt I am aware of provides a solution.

I will not consider Garrett’s attempted solution to the argument from attributes. Most importantly, we can’t find any reason to think the argument from affections is valid, so the proposition has not been established. This is not to say that the Leibniz objection is necessarily decisive, I shall just not consider any attempts at an answer here. Indeed, Garrett’s attempt isn’t obviously false, but I wish to note that it’s far from simple, whereas Spinoza seems to think the argument is obvious. As such, I have my doubts about Garrett’s attempt as an interpretation of Spinoza’s thought. Therefore, I would prefer to look for an alternative instead of examining this other solution. I will now turn to 1p8schol2 for an alternative argument for 1p5.

Spinoza argues as follows. Let us suppose a fixed number, say twenty, of individuals exist (i.e. more than one substance of the same nature). Spinoza is a causal rationalist: for everything that exists there needs to be a cause, through which we conceive the effect i.e., we need a cause for the existence of each individual. Spinoza also thinks that a true definition of X expresses only the nature of X. Therefore, no true definition expresses a fixed number of individuals: the definition of a triangle states that it is a 3-sided shape, not a fixed number of triangles. So, the cause of the existence of the twenty individuals cannot be the nature of the individuals (the nature can’t state twenty exist). If the cause isn’t in the nature of a thing, it must be in something else, by elimination. But to say the cause is external to the individuals is to say they aren’t ontologically independent: they aren’t causally self-sufficient (self-caused). This contradicts what it is to be a substance (1p6), so these twenty individuals cannot be substances. This argument clearly applies generally, so Spinoza concludes that there cannot be two substances of the same nature. The proposition is established.

What are we to make of this? Well, I think there is a problem. A substance is, for Spinoza, necessarily existing. Its essence involves existence (1p7). Spinoza’s demonstration of this is simple: no substance can be caused to exist by anything external to it, as if it were it wouldn’t be ontologically independent. But it needs a cause to exist, so it is self-caused. To be self-caused is just to exist necessarily (1Def1). Now, say we have twenty substances of the same nature. We need a cause for the existence of each, due to Spinoza’s causal rationalism. But, each substance has the same nature, and if it is a substance then its nature is such that it involves existence: it is necessarily existing. The nature which all twenty share doesn’t cause twenty to exist, but given we have twenty existing, we can explain the existence of each. The existence of each is just explained by the fact that each is a substance, and substance necessarily exists. Now, maybe we could counter: why do twenty exist in the first place? Surely causal rationalism demands an answer to this. But, for that matter, why should one substance exist? What is different about saying twenty substances exist and saying one does? In both cases we can explain the existence of each substance by its nature as substance, in neither case can we explain why we have the number we do. There does seem to be one possible objection here. Given substance necessarily exists, one substance of a particular nature must exist. So this explains the existence of one substance. But the essence, as stated, says nothing about a number greater than one. So there is a cause for one substance, but not for twenty. However, we can counter here too. Given that substance by its very nature exists, what stops twenty substances existing? Nothing external to the substances can, as this contradicts their nature as substances. As their essence isn’t self-contradictory (as a square triangle would be) nothing can stop them existing. Given nothing stops twenty substances existing, they exist. It seems to me reasonable to conclude that this demonstration of 1p5 fails. Now, my above argument looks like it would allow us to show as many substances of the same nature as we want exist. This is true, but I contend that I have argued above using principles Spinoza holds. The problem is with Spinoza’s notion of substance, not my argument.

Of course, Spinoza may well have other reasons for holding 1p5. I shall end this essay by considering what they could be, having shown that neither of his official demonstrations succeed. What I wish to say here shall be sketchy, but I believe I can point in the right direction . I wish to show why 1p5 might at least make sense in the context of Spinoza’s system, rather than conclusively demonstrate it. I shall focus on extended substance, and forget about thought . As I have said earlier, substances cannot enter into causal relations with each other (1p6). Not only can one substance never be the cause of another but also, if X causes Y, then for Spinoza, conception of Y requires conception of X (1Ax4). But substance is epistemically independent i.e. requires the conception of nothing else. So, in principle, no two extended substances can causally interact. So, however many extended substances we have they are completely independent of each other. But what is the prime candidate for the name extended substances (if we want more than one)? Bodies. But causally interacting with each other is one of the main characteristics of bodies. Indeed, the concept of substance was used historically to explain what exactly remains unchanged in such interactions. I contend that to say we can have more than one extended substance leads, within Spinoza’s system, to a nonsensical conclusion: that they cannot interact at all. If we only have one extended substance we have no problem, as there isn’t anything to interact with. Consequently, it seems we have good reason to believe that there cannot be two substances of extended nature. Of course, 1p5 is more general than this, and I am not sure if any version of the above can work with Thought. Also, Spinoza does seem to hold that there are other (unknowable) attributes, so this will cause further problems. My conclusion here is just that we are justified in asserting a limited version of 1p5: there cannot be two extended substances.

Of course, the above argument appeals to certain principles (mainly that substances can’t causally interact) that Spinoza accepts but others may not. My aim is not to show that Spinoza is right, just that there are good reasons to accept the limited version of 1p5, based upon Spinozistic principles. However, as indicated I am unsure as to whether a version of this argument can apply for other attributes. Certainly the idea of thinking substances (minds) being unable to causally interact poses no obvious problems. Our best hope maybe lies in the Parallelism that exists between the attributes, but there is no room for any development along these lines here. To conclude, in this essay I have demonstrated why both of Spinoza’s official demonstrations of 1p5 fail. I have then gone on to suggest a weakened version of 1p5 which we may think Spinoza has good reasons for holding. Whether we can extend our argument here I am unsure. A real problem will be that 1p5 occurs at such an early point in The Ethics that the Geometrical Method just won’t allow taking later propositions, such as the mentioned Parallelism, in order to provide support. As such major revisions to the order of Spinoza’s thought are needed.

Bibliography

Spinoza - The Essential Spinoza (ed. Michael L. Morgan, trans. Samuel Shirley, Hackett 2006)
Nadler, Steven – Spinoza’s Ethics (Cambridge University Press 2006)
Curley, E.M – Spinoza’s Metaphysics (Harvard University Press 1969)
Allison, Henry – Benedict de Spinoza: An Introduction (Yale University Press 1987)
Bennett, Jonathan – A Study of Spinoza’s Ethics (Hackett 1984)
Bennett, Jonathan – ‘Spinoza’s Metaphysics’ in Cambridge Companion To Spinoza (ed. Garrett, Cambridge University Press 1995)
Garrett, Don – ‘Ethics 1p5: Shared Attributes and the Basis of Spinoza’s Monism’ in Central Themes in Early Modern Philosophy (Eds. Cover an d Kulstad, Hackett 1990)
Descartes, Rene – Selected Philosophical Writings (trans. Cottingham, Stroothoff and Murdoch, Cambridge University Press 1988)

Article Source: http://journal.ilovephilosophy.com

Please Rate this Article

 

Not yet Rated

Click the XML Icon Above to Receive Philosophy Articles Via RSS!
Additional Articles From - Home | Philosophy
Sign Up as an Author.
Submit Articles
Member Login
Top Authors
Most Popular Articles
Submission Guidelines
Ezine Notifications
Article RSS Feeds
New Stuff
Forum
Link to Us
Contact Us

Powered by Article Dashboard