First Part of Spinoza’s Ethica: Concerning God

[ Page 2 ] [Page 3 ] [Page 4] [ Page 5 ]

Spinoza did not believe in a transcendent anthropomorphic god. He did believe that many of the proofs purportedly about god that he found in Descartes and, beyond Descartes, in the Scholastics were valid. He agreed that the Cartesian Cogito was a valid proof. He saw that the method of reasoning in the Cogito was nearly as similar to the method of proof in the ontological proof of the existence of god as two independent syllogistic deductions are similar to each other. He bought the Anselmian and Cartesian proof of the existence of god as well as Anselm's further proof that there is only one god and that god was the el supremo dictator of the universe. The problem he addressed in the First Part of the Ethics was how to comprehend both the validity of the proofs and the falsehood of the conclusions.

Much of the gibberish of the First Part of the Ethics is systematic. It arises not only from the real meaninglessness of the scholastic terminology Spinoza utilized but also from his misguided attempt to cram the type of reasoning that appeared in the ontological proof and that Descartes applied with such éclat to a novel subject with his Cogito, a type of reasoning that by its very nature cannot be made deductive, into the pseudo-logical consequentiality of an axiomatic deduction that has only the vaguest similarity to the geometrical method Spinoza prizes. We may surmise that the gibberish is also partly rhetorical. Not a word in the Ethics contradicts an interpretation that the proofs therein are no more than a fairly traditional set of proofs about the Xtian concept of god. Even the famous phrase “deus sive natura” has a kissing cousin in the Meditations although Descartes unequivocally subscribed to the Xtian concept of god and would certainly have abjured those anti-theistic remarks found in the Tractatus and elsewhere that allow us to piece together the true intent of the Ethics.

The infelicities of the First Part of the Ethics have specific causes. The first is his misunderstanding that the ontological proof of the existence of god could be recast in deductive form. Spinoza could not assemble into an acceptable set of propositions the challenge, necessary for the proof, to conceive of a most perfect entity (Nor, notably, could Descartes in his own failed attempt in the Second Replies). What he does deduce, or try to deduce, is the identity of a cause of itself with something that is conceived through itself. This “proof” is nothing more than a series of equivalences by definition and so is not a proof at all but a word game. The actual ontological proof and minor variations thereof are relegated to various notes and addenda to the propositions. The other cause is much more intriguing. Read one way, as we said, the propositions of the First Part of the Ethics may be taken as an unimpeachable restatement of the Anselmian proofs of the existence, oneness etc. of a personal god. However, this reading depends on understanding god as a substance separate from his creation, which in turn is composed of many other substances. Spinoza unhinges this reading by subtly turning the proof that there is only one god (a standard move in monotheistic theology) into a proof that there is only one substance. If we understand “substance” in the same way as Anselm and Descartes, then this is nonsense.  But if “substance” is understood somewhat differently as something on the order of “our world,” a name for everything around us including our thoughts and ourselves as well as the physical universe, then the proof that there is only one substance is not a monotheistic gambit. It is rather a proof that there is no anthropomorphic god at all. Spinoza covers his tracks in two ways, first by constantly using the term “god” interchangeably with “rerum natura” and “substance,” and, secondly, by brandishing the lack of a definite article in Latin to produce real, and perhaps intentional, confusion between “the substance that is the one god” and “substance, which happens to be all there is” (It is this deft maneuver that led a puzzled Hume to construe Spinoza’s “substance” as "substrate"). The First Part of the Ethics is what Freud would have called an “absurd dream” that was verworren by a censorship made up of equal parts real danger, the desire not to appear as an atheistic pamphleteer and a concern with the philosophical issue of creating a conceptual structure more sensitive to the world measured by Galileo and a rational morality subsequent to his Umwertung aller theologischen Werte.

The First Part of Spinoza’s Ethics contains proofs of the following three primary arguments (The other propositions in the First Part are subordinate to these broad conclusions. They are either intermediate steps in the proof of the final conclusions or consequences derived from the truth of those conclusions):

I. Substance exists necessarily.

II. Only one substance exists.

III. god is substance. Substance is god.

Spinoza does not prove a single one of these conclusions either logically or according to his self-professed standard of more geometrico. Almost every step in Spinoza’s reasoning is infected with some logical error or another.

However, as with many of Spinoza’s philosophical points, he is really making the case for a much more concrete conclusion. He often approaches that conclusion from two different directions, first from an extremely abstract logical/geometrical/metaphysical point of view and then from the angle of what might be called a real world argument. In the case of the First Part of the Ethics Spinoza’s concrete conclusion is that there is no transcendent or anthropomorphic god. The main body of the First Part of the Ethics consists in his geometrical proof. The famous Appendix contains his real world argument. In almost every instance the geometrical proof is an abysmal failure and the real world argument is genuinely convincing. If you can’t live happily ever after with Pamela Anderson, then paying for a fairly decent hooker is not a bad substitute. Or, to paraphrase Mick, if you can’t get what you want make do with what you can.

Here are Spinoza's three broad conclusions followed by the geometrical arguments in their favor and the objections to the validity of those arguments.

I.Substance Exists Necessarily:

1. A cause of itself cannot be conceived except as existing (Def. I)

2. Substance is conceived through itself only. Its conception does not depend on the conception of another thing. (Def. III and Axiom II)

3. One substance cannot be produced (caused) by another substance for its conception would then depend on another thing. (Prop. VI)

4. Therefore substance must be a cause of itself. (Prop. VII)

5. Therefore existence belongs to the essence of substance. (Def. I and Prop. VII)

6. From a given cause its effect follows necessarily; without a cause there can be no effect. (Axiom III)

7. Since substance causes itself, its essence (existence) follows necessarily. Therefore substance exists necessarily. (Prop. VII)

A subordinate step in the argument between (2) and (3) is never quite explicitly stated although it seems to be implied in Prop. III:

                (2a) If something is caused by another thing, then its conception is dependent on the other thing.


I leave aside the time-honored objection that existence is not a predicate (Kant). Even though this objection is valid (E.g., assuming Spinoza does not mean that “Substance exists” is somehow equivalent to “Substance can be found somewhere in time and space”, what then can he mean? Does substance exist or exist necessarily just because we string together these words?), nevertheless the specific error in Spinoza’s proof lies in not showing that it follows that, if something is not caused by something else, then that something is caused by itself and furthermore that that something exists and necessarily exists. Spinoza appears to escape Hume’s objection that causality is not an a priori relation because his concept of causality muddles logical dependence and physical causality. However, the mechanism by which Spinoza avoids that problem regarding causality leads other problems as will be seen below.

1. Def. I may be empty, i.e. we may not be able to conceive a cause of itself. Likewise we may not be able to conceive of something whose nature involves existence.

3-4. First Objection. Even though substance cannot be conceived as caused by something else, it does not follow that substance exists. Where "a" and "b" stand for "things" (We will leave "thing undefined") and a≠b, Spinoza’s penultimate conclusion is:

(A) There is something (a) such that for every other thing (b) it is the case that, if it is not the case that b causes a, then it is the case that a causes a.

So, since by Def. 1 a cause of itself cannot be conceived except as existing, then if the concept of a can be conceived, then a exists. This is not a logically valid proof because the following statement is not a contradiction:

(B) It is not the case that b causes a and it is not the case that a causes a.

Can Spinoza’s conclusion be saved by recasting "a causes a" as a tautology (which would be implied by any statement, true or false)? In that case the equivalent of “a causes a” would be “If a exists then a exists.”  Then Spinoza’s conclusion would read:

(C) It is the case that, if it is not the case that if b exists then a exists, then, if a exists then a exists.

Obviously not,  because “If a exists then a exists” also follows from “If b exists then a exists.” In other words, “If b causes a then a causes a.” For (C) is logically consistent with:

(D) It is the case that, it is not the case that if b exists then a exists, and it is not the case that a exists.

One might argue that Spinoza’s true understanding of the logical dependence of the existence of something following from its definition reads as follows:

(E) If a can be defined as existing then a exists.

In order to maintain the symmetry between “b causes a” and “a causes a” the If clause would have to be written either as “If b can be defined as existing…” or else “If b can be defined such that a exists…” Under the first alternative (C) would read:

(F) It is the case that, if it is not the case that if b can be defined as existing then a exists, then, if a can be defined as existing then a exists.

This does not make any progress for (F) is logically compatible with:

(G) It is not the case that a exists.

Under the second alternative (C) would read:

(H) It is the case that, if it is not the case that if b can be defined such that a exists then a exists, then, if a can be defined as existing then a exists.

Clearly (H) is also logically compatible with (G). (F) and (H) in fact are only well formed as long as we do not try to interpret "defined as existing" or "exists". In that case (F) says no more than (where p, q, r and s stand for arbitrary propositions):

(I) If (Not (if p then q)) then (if r then s)

With a little work this can also be restated more precisely and retaining the references to a and b using predicate calculus. However, that is not required. (I) would remain simply the form of a proposition, not a logically necessary inference.

However ill formed it may be, (E) is the closest restatement in clearer language of the concept of a thing whose definition involves existence. Showing the fault in (F) and (H) involves more than just the non-predicability of existence since Spinoza's argument involves more, namely that if the existence of a thing is not caused  by something else, then that thing must be defined as existing and therefore exists.

3-4. Second Objection. Just because some thing, b, does not cause some other thing, a, it does not follow that a causes itself. a  might just be there, hanging around so to speak. Similarly, from “the concept of a is not logically dependent on the concept of  any b” it does not follow that “the concept of a is logically dependent on the concept of  a” except insofar as “the concept of a is logically dependent on the concept of a” follows from any statement true or false. Some have argued that giving up the idea that there may be something, namely substance, that has no cause, is to give up the idea that the universe can be rationally explained. Nonsense. A scientific theory does not look for an efficient cause for the phenomena it was created to explain taken as a whole though it may provide a framework of causal relations for interactions within the set of phenomena about which the theory has been created. In fact from a scientific point of view it might be better to say that each phenomenon has not so much a cause as a history. A theory might be characterized as a summary of observed regularities based on a few principles. Indeed the research imperative that we seek a reason for individual things does not extend to seeking a reason for everything taken as a whole. Numerous philosophers and scientists consider the latter quest to be “speculative” and “metaphysical.”  Moreover, I know of no mathematical physics that even has a concept of something causing itself. If we give up the idea of transcendent causality, as Spinoza does, what’s wrong with saying the universe is just there? (This also affects the second half of Axiom III since something may be without being an effect.)

3-4. Third Objection. Just because the concept of some thing or class of things, a, is not “conceptually dependent” (“Conceptual dependence” is undefined. Does it have something to do with definitions or some other relation? Spinoza doesn’t tell us) on the concept of some other thing or class of things, b, it does not follow that b does not cause a. Presumably b may cause a even if a definition or understanding of b is not required in the definition or understanding of a. We may produce a more or less adequate definition of “Neanderthal” without including in that definition the concept of parenthood or mentioning the parents of each Neanderthal. If we demand that an adequate definition include the definition of anything that could have some causal relation to a, then every concept would define every other concept and there would be no understanding of either a or of conceptual dependence at all. If we argue that, by saying that “substance” is not conceptually dependent on the concept of any other thing, Spinoza is simply highlighting its role as a kind of logical placeholder, then “substance” would not stand for anything in particular, certainly nothing that could enter into specific causal relations with itself or anything else. Furthermore, if "a causes a" has any meaning at all aside from logical equivalence (“Cause” is also undefined), there is no logical reason why "It is not the case that b causes a" and "It is not the case that a causes a" cannot both be true. But if "a causes a" means no more than logical equivalence between the concept of a and itself, that is if it means no more than a=a, then the statement of that equivalence is a tautology and "a causes a" is true whether or not "b causes a" is true or not true. In fact it is true of everything.

Isn't there something garbled in the very idea of something causing itself? Is there any difference between saying “a causes itself” and “a has no cause” The very concept of cause and effect seems to involve the non identity of cause and effect.

5. Either “essence equals existence” means something different from “cause of itself” or it doesn’t. In other words, either “essence involves existence” adds content to concept of a cause of itself or it does not.  If it does not add content then this is an empty definition. I might as well say by “cause of itself” I mean a “logjam turtle.”  If it does add content, then Spinoza’s argument that since substance is a cause of itself then its essence must involve existence is not a purely logical argument, i.e. it does not follow from the definition alone. Spinoza would have to provide independent definitions of “cause of itself” and “something whose essence involves existence” and then prove that everything defined by the first ("cause of itself") is also defined by the second ("something whose essence involves existence").

6. "From a given cause its effect follows necessarily": Either this is a relation of logical dependence in which case it is empty or else "b causes a" and "b necessarily causes a" mean different things in which case "necessarily" would have to be defined and a missing argument to the effect that a necessary causal relation follows logically from a causal relation would have to be supplied. "Without a cause there can be no effect": I imagine this is supposed to provide the missing step in the argument that if a thing is not caused by thing else it must cause itself. It does not succeed. Yes, the concept of an effect cannot be understood without the concept of its cause. But this does not mean that if an external cause, b, is missing for any a, then that a must cause itself. For a may not be an effect.

Perhaps by "necessary" Spinoza means things could not have been otherwise on grounds other than logical grounds. In that case substance exists necessarily means that things could not have been otherwise than that substance exists. This conclusion is palpably false. The only difference is Spinoza would not have been around to prove its existence.

The actual text of First Part of Spinoza’s Ethics contains four separate arguments for the existence of substance scattered haphazardly through his presentation.

The first appears to be no more than an attempt at a proof that if something is conceived through itself then it must thereby be a cause of itself. The actual existence of such a thing appears to be stipulated by definition since (I Def. 1) a cause of itself is defined as existing. This is the upshot of the principal proofs of I Props. VII and XI.

The second appears in I Prop. VIII Note 2. Here, in an almost literal transcription of one of Anselm’s proofs, Spinoza states that if we have a clear and distinct notion of something whose essence is to exist, then that something must exist: "...if anyone should say that he has a clear and distinct, that is a true, idea of substance, and should nevertheless doubt whether substance existed, he would indeed be like one who should say that he had a true idea and yet should wonder whether it were false..." The answer is the same: Perhaps we don't have and cannot have a clear and distinct or true idea of "substance" as Spinoza understands it. A Hume-style argument may find its proper place here. It would run as follows: The majority of scientists and even philosophers have been not been able to figure out what is meant by something whose essence is to exist. An empirical review of the history of science and philosophy shows that the preponderance of the evidence weighs against a comprehensible definition. Whereas a point, namely a thing with location but not extension, has become clear and meaningful through much usage by mathematicians and scientists, no one has been able to put the entity whose essence is to exist into an experiment or a geometrical figure. Spinoza is not successful because of the logical faults in his deductions.

The third appears in I Prop XI Another Proof 1. These are the steps of Spinoza’s argument: (1) The existence of anything that exists must have a cause. Likewise, the non-existence of anything that doesn’t exist must have a cause. (2) If something exists it exists necessarily. If something does not exist, it is impossible for it to exist. (3) If there is no reason (cause) that something does not exist then that thing must exist necessarily. (4) There is no reason (cause) why substance (god) does not exist because: (a) The cause is either part of substance or not part of substance. (b) If the cause is part of substance then substance exists because the cause exists. (c) If the cause is not part of substance then it can have no causal relation with substance by I Prop. 2 and so it cannot be the cause of the non-existence of substance (god). This is really terrible. First, no argument is presented that there must be a cause of the existence or non-existence of things and the claim that non-existence needs a cause is just absurd. Secondly, (2) might follow from (1) by the Law of Excluded Middle, but lacking a definition of “necessary,” that consequence has not been established. The same applies to the derivation of (3) from (2). Third, assuming that (1), (2) and (3) are true or even meaningful then, by modus tollens, if substance does not exist then the cause of the non-existence of substance is neither part of or outside of substance since there is nothing for it to be part of or outside of.

An interesting comment appears in this passage regarding the existence of geometrical objects. Existence is not part of their definition as is the case with substance. Rather their existence comes from the order of universal corporeal nature (“ex ordine universæ naturæ corporeæ”). This comment seems to endorse a rather anti-Pythagorean view that geometrical objects do not maintain a separate existence. Instead their existence in Spinoza’s mind seems to be tied to the way we study and measure nature.

The fourth appears in I Prop XI Another Proof 2. This is duplicates Anselm’s principal ontological proof, the only difference being that Spinoza substitutes “powerful” (potentia) for Anselm’s “best” (optimum). “Best” is actually better than “powerful” for, as we all know, absolute power corrupts absolutely. For some completely mysterious reason Spinoza calls this proof a posteriori.

Note 1: Spinoza's conception of substance as something whose conception depends on no other thing (alterius rei) mirrors Anselm's one thing that exists through itself (ipsum solum per seipsum). The logical consequence of this definition, as we shall see, is that there is nothing distinct from substance defined in this way. Everything else is just a mode or an attribute of this substance. The upshot is nothing more than a mockery of the ontological proof of the existence of god by showing that it merely proves that something, dubbed "natura," exists.

Note 2: Spinoza uses the term "thing (res)" as an undefined primitive (I Def. II, I Def. VII, I Prop. III, I Prop. IV etc.). Later (Book II, Prop. XL, Note 1) he complains that "res", like "being" and other transcendental terms results from a confused attempt to generalize from images formed in the mind of individual things. He calls it a "universal attribute." But he does not explain how his use of "res" as a primitive term in the First Part escapes this criticism.

Spinoza does not use "universe" as an undefined primitive (Prop. V). Elwes/Boyle/Parkinson's "universe" is a misleading translation of "rerum natura."

"Natura" is an undefined primitive and thereby constitutes a serious flaw in Spinoza's exposition particularly since he uses "natura" in more than one way, e.g.. in the phrase "id cujus natura" (Def. I), and in the phrase "rerum natura" (Prop. V) and in later the phrase "Deus seu natura" (Fourth Part, Preface). The last is such a celebrated phrase that it is a shame that Spinoza did not clarify the ambiguity. [ Next ]

[ Page 2 ] [Page 3 ] [Page 4] [ Page 5 ]