The Cosmological Argument
There are proofs for God’s existence that are loosely grouped together as variations of the Cosmological Argument, and this is where R. Kelemen finds his next argument for God’s existence. They generally argue that some being must have created the universe or causes it to function, and this being is God. Arguments along these lines can be traced back to Aristotle and his concept of an Unmoved Mover. In Jewish texts, there doesn’t seem to be much discussion of this approach until the Medieval period (see this post). Dr. Herbert Davidson wrote an excellent book, Proofs for Eternity, Creation and the Existence of God in Medieval Islamic and Jewish Philosophy (NY, Oxford University Press: 1987), in which he traces the proofs through Medieval muslim and Jewish sources.
The Kalam Cosmological Argument
The first phase of Cosmological Arguments we will discuss is the Kalam Cosmological Argument. There are actually many of this kind but the general approach is to prove that the universe and everything in it was created using the premise that there cannot be an infinite chain of creators going back in time. From this it emerges that there was a First Cause that initially created the universe.
Let’s look at a Jewish version of this argument, that of R. Bachya Ibn Pakuda (Chovos Ha-Levavos, Sha’ar Ha-Yichud chs. 5-6). R. Bachya’s argument proceeds like this (albeit stripped of most of the detail):
1) Something that exists could not have made itself. [cf. R. Sa'adia Gaon, Emunos Ve-Dei'os 1:2]
2) An infinite regress is impossible; something that exists cannot trace its cause backwards in time infinitely because then it would have no start and would not currently exist. [cf. Emunos Ve-Dei'os 1:1, fourth proof]
3) There must have been a First Cause to begin the chain of creation.
R. Bachya concludes by arguing against those who say that the universe came into existence by chance. He does this with a Teleological Argument, which we will discuss at a later time.
The Kalam Cosmological Argument’s assumptions are questionable. Later philosophers (e.g. Rambam) denied that we can prove that the universe was created. However, contemporary science seems to imply a beginning â€“ the Big Bang â€“ and therefore we may not need to prove this point philosophically. It is with the aid of contemporary science that the Kalam Cosmological Argument has been revived in the past few decades, most notably by William Craig. The main point he has to prove is that an “infinite regress”, a series that retreats infinitely without an origin, is impossible. While he has made a strong case for this, most philosophers disagree with his conclusion about a chronological infinite regress, and allow for it in various cases that undermine his argument for a First Cause.
The Maimonidean Arguments
Later medieval philosophers rejected the Kalam enthusiasm for proving the creation of the universe as philosophically naive. Instead, they tried different approaches. In particular, the Rambam proposed what is called a constructive dilemma (or in Talmudic terminology, mi-mah nafshakh): If the universe was created then God exists because of the Kalam arguments and if the universe was not created then God exists because of the following arguments. Either way, he’s proven his point (Moreh Nevukhim 1:71).
The philosophical literature tends to focus on Thomas Aquinas’ proofs, mainly because Christians dominate the field. It seems to me worthwhile to present the Rambam’s proofs rather than Aquinas’, particularly since they are so similar (Aquinas not only read the Rambam but quoted him). However, since the Rambam’s proofs as detailed in Moreh Nevukhim (2:Intro-1) are extremely involved (he begins with 26 propositions), we can only present the headlines as we did above for R. Bachya.
I. The first argument was originally formulated by Aristotle (see Davidson, ch. 8) and argues from the phenomena of motion and change. What causes them to happen? This is not a question of priority in time, like a bowling ball in motion was previously rolled by a person. Rather, it is a matter of priority in cause, like a fire heats a pot and the pot heats a stew. According to Aristotelian science, everything in motion must have a simultaneous cause which eventually leads back to the celestial sphere. The celestial sphere itself has a cause, and that is the First Mover. Here is the proof:
1) Everything that is moving must have a mover.
2) An infinite regress is impossible.
3) There must be a First Mover who starts the chain of movement.
Modern science teaches us the concept of inertia, which means that something in motion need not be constantly moved (cf. R. Nachum Rabinovitch, Yad Peshutah, Hilkhos Yesodei Ha-Torah 1:2). Additionally, as R. Chasdai Crescas (Or Hashem 1:2:15) points out, an object can move by its own nature and not because of an external mover. Together, these two objections imply that, for example, a planet can revolve around the sun eternally because that is the nature of the planet. However, as we will see, there is still a way to reframe this proof for the contemporary understanding of the natural world.
II. The second argument is also towards a First Mover but avoids the issue of infinite regression. The core of the argument is the following Aristotelian principle we will call P1: If two characteristics can be found in an object and one of those characteristics can be found alone in another object, then the other characteristic must also be found alone in an object. For example, from the fact that honey-vinegar exists and honey alone exists, we can deduce that vinegar must exist on its own as well.
Here is the basic argument:
1) Most objects we see are both moved by something and move other things.
2) Some objects do not move others but are moved (i.e. are a Final Mover).
3) Based on P1, if the characteristic of being moved can be found alone then there must exist an object that moves but is not moved, i.e. if there is a Final Mover then there must be a First Mover.
R. Yosef Albo (Sefer Ha-Ikkarim 2:5) states that this argument is weak because P1 seems to be untrue. Just because you find a pellet that is red and oval, and there are other red objects, does not mean that there must be oval pellets that are not red. R. Chasdai Crescas (Or Hashem 1:2:16) agrees that the principle is incorrect and therefore the proof does not work.
III. The third argument tries to prove a Necessary Being in a negative way: If there was no Necessary Being then nothing would exist. It proceeds as follows:
1) Objects have either finite lives or infinite lives.
2) If something has a finite life then there is some time when it does/did not exist.
3) If everything has a finite life then there was a time in the past when nothing existed.
4) If there was a time in the past during which nothing existed, then nothing could exist now.
5) Therefore, there must be a being with an infinite life that always exists (a Necessary Being).
Argument 3) seems easily refuted. Why can’t there be an infinite series of overlapping finite lives? A lives for 100 years; B is born when A is 90 and lives 100 years; C is born when B is 90 and lives 100 years, etc. forever. Even if each object goes out of existence at the same time, it does not follow that they all go out of existence at the same time. This is a point that R. Chasdai Crescas makes (Or Hashem 1:2:17).
Additionally, as the Rambam himself says, this proof assumes an infinite past. If not, there is no reason that every possible scenario took place in the past and argument 4 does not follow argument 3. Really, there is an assumed argument 3b) that the past is infinite.
However, with this understanding we can restate arguments 3) and 3b) to avoid Crescas’ objection:
3) If everything has a finite life then there is a possible scenario in which nothing exists.
3b) The past is infinite and over an infinite stretch of time every possible scenario is recognized (cf. Davidson, p. 382).
However, it is not at all clear that argument 3b) is true. As Alvin Plantinga (God and Other Minds, ch. 1) shows, there is no wat to satisfactorily restate 3b) to be true in all cases.
IV. The fourth argument is about the transition from the state of potential to actual. There is a constant movement from potential to actual, e.g. something that can become hot actually becomes hot. According to Aristotle, only something already hot, such as a fire, can make an object become actually hot. But what brought the fire into actuality? If you trace back all of the causes, you end up at a First Cause. Here is the proof:
1) Anything that moves from potentiality to actuality requires an outside cause.
2) An infinite regress is impossible.
3) There must be a First Cause.
As above, the impossibility of an infinite regress is debatable. R. Chasdai Crescas (Or Hashem 1:2:17) also points out that with this argument, there could be multiple first causes. Beyond that, though, there are multiple questionable assumptions underlying this argument â€“ e.g. that something cannot bring itself from potentiality to actuality.
V. R. Chasdai Crescas (Or Hashem 1:3:2) offers another argument that I think is worthwhile to include in this list. It is similar to the Rambam’s first and fourth arguments but avoids some of the problems he raised with them. R. Crescas argues that anything that exists requires a cause that brought it into existence. This is not the same as the Kalam argument because it is not referring to cause in the specifically chronological sense but in terms of causal or logical priority (like a hand moves a golf club and the club moves a ball).
Here is the proof:
1) Everything has a cause.
2) An infinite regress is impossible.
3) There must be a First Cause.
To sum up these medieval proofs:
1. Kalam – argues for a First Cause that is a creator
2. Rambam 1 – argues for a First Mover
3. Rambam 2 – argues for a First Mover (from a Final Mover)
4. Rambam 3 – argues for a Necessary Being (that always exists)
5. Rambam 4 – argues for a First Cause (from the actualization of potential)
6. Crescas – argues for a First Cause in terms of logical priority
As an aside, Josef Stern has an interesting discussion on the relationship between Rambam’s four proofs in his “Maimonides’ Demonstrations: Principles and Practice” in Medieval Philosophy and Theology 10 (2001).
AAs we said earlier, most discussion of the Cosmological Argument focuses on Aquinas. It might be helpful right now to stop for a moment and point out the correspondence between the above arguments and Aquinas’ Five Ways. Rambam 1 & 4 are encompassed in Aquinas’ first way. Crescas’ argument is similar to Aquinas’ second way. Rambam 3 corresponds to Aquinas’ third way.
Until now, we have reviewed the medieval cosmological proofs. Most of this discussion can be dismissed because it is based on outdated understandings of the physical universe. However, as we move into the Modern Era, we will see that we can still find meaning in the arguments.
Leibniz is often credited with ushering in a new era of the Cosmological Argument. His argument is based on what is called the Principle of Sufficient Reason. This principle states that every fact has an explanation. While this is a difficult principle to prove, it seems to be the basis of all rational inquiry into the universe. Leibniz is generally credited with bringing this principle into the discussion of the arguments for God’s existence, although intellectual historians will tell you that it is implicit in the earlier arguments and that Spinoza had already used the principle in this area. Be that as it may, armed with this principle, Leibniz’s argument can be simplified as follows:
1) Anything that exists must have a cause that explains its existence.
2) The universe exists.
3) There must be a being whose existence is self-explanatory and can explain the existence of the universe.
This is a powerful approach to the subject because it leaves behind ancient and medieval physics about causes and movers. To some degree, it sets aside all physics and focuses on the limits of science. It is often said that science can explain “how” but it is up to religion to explain “why”. Leibniz points out that there is one “how” that science cannot explain â€“ the first “how”.
We can also restate many of the medieval arguments in this way. For example, we are looking for an explanation of motion or an explanation of change. Regardless of our understanding of physics, we can still search for an explanation of how motion and change originally began. What can explain the beginning of the chain of motion/change?
To summarize what we have seen until now, the Kalam argument looks for a first creator, the Rambam’s first and second arguments look for a first mover, his third argument looks for a necessary being, his fourth argument and Crescas’ argument look for a first cause and Leibniz’s argument looks for a first explanation.
The two most devastating critiques of the Leibnizian Cosmological Argument come from David Hume and Immanuel Kant. Hume, and more explicitly Kant, deconstruct the Cosmological Argument into an offshoot of the Ontological Argument. The Ontological Argument, which we will discuss at a later time, attempts to prove that God exists because He is a logically necessary being; it is logically impossible for Him not to exist. Hume and Kant contend that the Cosmological Argument ultimately requires proving that God is a necessary being and therefore exists (objection 1).
Kant points out that there is an unwarranted step in which a necessary being is equated with a most-perfect being, at which point we can deduce attributes of God. This equation has no basis and, therefore, there is no way to know anything about God from a First Cause (objection 2). <
Hume also questions why we keep looking back for further causes, movers or explanations until we reach God, at which point we stop looking back. Why don’t we look to find who caused, moved or explains the existence of God? Or, if we are going to stop somewhere, why not stop at the universe as the First Cause, etc. (objection 3)?
However, as Davidson (pp. 399-404) points out, while these objections achieved great influence, it is not clear that they are decisive (David Conway systematically addresses objections in his The Rediscovery of Wisdom). Leibniz’s argument does not require the concept of a necessary being to succeed. What it uses is the idea of a self-explanatory being, something that exists by virtue of itself. This being need not be logically necessary, i.e. he may or may not exist but if he does, his existence is by virtue of himself and not something else (response to 1).
It is not clear that there is any need to equate a necessary being, or really a being whose existence is self-explanatory, with a most-perfect being. There are ways of deducing qualities of the being whose existence is self-explanatory, and even without that the argument can be successful without offering a detailed description of the First Cause whose existence is proven (response to 2).
The third objection is seriously undermined by scientific developments of the twentieth century. The rise of the Big Bang Theory provides an explanation of why we ask who created the universe and do not stop with it as the First Cause: The universe gives every appearance of having been created. While there are theories of the universe’s beginning that do not involve a creator, the need to explain how the universe came into being opens the door for questioning what caused the universe’s existence.
Recall that a significant reason for the rejection of the Kalam argument was that it includes a proof for the creation of the universe, an argument which the Rambam and others rejected as inconclusive. However, with the ascendance of the Big Bang Theory, we no longer have to prove that the universe came into being because science allows us to assume it as a fact. With that premise, some (as discussed above) have resurrected the Kalam Cosmological Argument â€“ since the universe came into being, there must have been a First Cause.
R. Kelemen steps into the discussion here. He points out that the Big Bang theory indicates that the universe was began at some point, and this implies the existence of God. This is not entirely correct because there are other cosmological theories of how the universe could have come into existence. However, the alternative theories of what preceded or caused the Big Bang seem to many to be extremely unlikely (see Antony Flew, There Is A God, pp. 137-143). As Richard Swinburne writes (The Existence of God, p. 152):
There is quite a chance that, if there is a God, he will make something of the finitude and complexity of a universe. It is very unlikely that a universe would exist uncaused, but rather more likely that God would exist uncaused. Hence the argument from the existence of the universe to the existence of God is a good… argument.
Swinburne makes the probabilistic argument that he uses in many other contexts: It seems more probable that if there is a God that He would create the universe than if there is no God that the universe would come into existence. Therefore, the existence of the universe is an argument for the likelihood of God’s existence. Put in different words, God is a better explanation for the existence of the universe than any other theory. This is the type of probabilistic argument we discussed in an earlier post (link), and a Jewish philosopher — George Schlesinger — has written a good deal along this line of thinking.