The Four Causes

What are there four of?

  1. Aristotle’s doctrine of the four causes is crucial, but easily misunderstood. It is natural for us (post-Humeans) to think of (what Aristotle calls) “causes” in terms of our latter-day notion of cause-and-effect. This is misleading in several ways:

    1. Only one of Aristotle’s causes (the “efficient” cause) sounds even remotely like a Humean cause.
    2. Humean causes are events, and so are their effects, but Aristotle doesn’t limit his causes in that way. Typically, it is substances that have causes. And that sounds odd.

  2. But to charge Aristotle with having only a dim understanding of causality is to accuse him of missing a target he wasn’t even aiming at. We must keep this in mind whenever we use the word “cause” in connection with Aristotle’s doctrine.

  3. We will begin with the question, What is it that Aristotle says there are four of? The Greek word is aition (plural aitia); sometimes it takes a feminine form, aitia (plural aitiai). And what is an aition? Part of Aristotle’s point is that there is no one answer to this question. An aition is just whatever one can cite in answer to a “why?” question. And what we give in answering a “why?” question is an explanation. So an aition is best thought of as an explanation than as a cause.

  4. Even so, that’s not enough. First, Aristotle thinks that you can ask what the aitia of this table are, and it’s not clear what sense, if any, it makes to ask for an explanation of the table. Second, he thinks that, in some sense, a carpenter is an aition of a table, and it’s not clear in what sense, if any, a carpenter (or anything like a carpenter) could be an explanation of anything.

    Here perhaps Ackrill’s “explanatory factor” is a more illuminating translation of aition. That is, an aition is something that plays a role as an explanatory factor in the explanation of something. But, as we’ll see, there are many kinds of explanations.

Where to find the doctrine in Aristotle’s texts

  1. In RAGP: Phys. II.3; and (extensively) in Metaph. A.3 ff. See also Part. An. 639b12ff
  2. Additionally (not in RAGP): APo. II.11; Metaph. Δ.2; GC 335a28-336a12.

The traditional picture

The picture is Aristotle’s, but the names of the causes are not. Quotations from Physics II.3, 194b24 ff:
  1. Material cause: “that out of which a thing comes-to-be and which persists is said to be a cause, for example, the bronze is a cause of a statue, the silver is a cause of a bowl, and the genera of these [is also a cause].”
  2. Formal cause: “the form or paradigm, and this is the formula of the essence … and the parts that are in the formula.”
  3. Efficient cause: “the primary starting point from which change or rest originates; for example, someone who has given advice is a cause, the father [is a cause] of a child, and in general what does [is a cause] of what is done and what alters something [is a cause] of what is altered.”
  4. Final cause: “[something may be called a cause] in the sense of an end (telos), namely, what something is for; for example, health [is a cause] of walking.”
This account makes it seem as if Aristotle is offering a catalog of causes, and is claiming that each thing has four different kinds of cause. But what the account misses is the idea that there is something ambiguous about the notion of aition.

The ambiguity of aition

Aristotle warns us of the ambiguity at 195a5: “causes are spoken of in many ways.” This is his usual formula for telling us that a term is being used ambiguously. That is, when one says that x is the aition of y, it isn’t clear what is meant until one specifies what sense of aition is intended:

  1. x is what y is [made] out of.
  2. x is what it is to be y.
  3. x is what produces y.
  4. x is what y is for.
This makes it hard for us to get clear on what Aristotle was up to, since neither “cause” nor “explanation” is ambiguous in the way Aristotle claims aition is. There is no English translation of aition that is ambiguous in the way (Aristotle claims) aition is. But if we shift from the noun “cause” to the verb “makes” we may get somewhere.

The ambiguity of makes

Aristotle’s point may be put this way: if we ask “what makes something so-and-so?” we can give four very different sorts of answer—each appropriate to a different sense of “makes.” Consider the following sentences:
  1. The table is made of wood.
  2. Having four legs and a flat top makes this (count as) a table.
  3. A carpenter makes a table.
  4. Having a surface suitable for eating or writing makes this (work as) a table.

Aristotelian versions of (1) - (4):

1a.  Wood is an aition of a table.
2a.  Having four legs and a flat top is an aition of a table.
3a.  A carpenter is an aition of a table.
4a.  Having a surface suitable for eating or writing is an aition of a table.

These sentences can be disambiguated by specifying the relevant sense of aition in each case:

1b.  Wood is what the table is made out of.
2b.  Having four legs and a flat top is what it is to be a table.
3b.  A carpenter is what produces a table.
4b.  Eating on and writing on is what a table is for.

Static vs. Dynamic Causes

Matter and form are two of the four causes, or explanatory factors. They are used to analyze the world statically—they tell us how it is at a given moment. But they do not tell us how it came to be that way. For that we need to look at things dynamically—we need to look at causes that explain why matter has come to be formed in the way that it has. Change consists in matter taking on (or losing) form. Efficient and final causes are used to explain why change occurs.

This is easiest to see in the case of an artifact, like a statue or a table. The table has come into existence because the carpenter put the form of the table (which he had in his mind) into the wood of which the table is composed. The carpenter has done this for the purpose of creating something he can write on or eat on. (Or, more likely, that he can sell to someone who wants it for that purpose.) This is a teleological explanation of there being a table.

This seems like a plausible doctrine about artifacts: they can be explained both statically (what they are, and what they’re made of) and dynamically (how they came to be, and what they are for).

Causes of natural objects

But what about natural objects? Aristotle (notoriously) held that the four causes could be found in nature, as well. That is, that there is a final cause of a tree, just as there is a final cause of a table. Here he is commonly thought to have made a huge mistake. How can there be final causes in nature, when final causes are purposes, what a thing is for? In the case of an artifact, the final cause is the end or goal that the artisan had in mind in making the thing. But what is the final cause of a dog, or a horse, or an oak tree?

  1. What they are used for? E.g., pets, pulling plows, serving as building materials, etc. To suppose so would be to suppose Aristotle guilty of reading human purposes and plans into nature. But this is not what he has in mind.

  2. Perhaps he thinks of nature as being like art, except that the artisan is God? God is the efficient cause of natural objects, and God’s purposes are the final causes of the natural objects that he creates.

No. In both (a) and (b), the final cause is external to the object. (Both the artisan and God are external to their artifacts; they impose form on matter from the outside.) But the final causes of natural objects are internal to those objects.

Final causes in nature

  1. The final cause of a natural object—a plant or an animal—is not a purpose, plan, or “intention.” Rather, it is whatever lies at the end of the regular series of developmental changes that typical specimens of a given species undergo. The final cause need not be a purpose that someone has in mind. I.e., where F is a biological kind: the telos of an F is what embryonic, immature, or developing Fs are all tending to grow into. The telos of a developing tiger is to be a tiger.

  2. Aristotle opposes final causes in nature to chance or randomness. So the fact that there is regularity in nature—as Aristotle says, things in nature happen “always or for the most part”— suggests to him that biological individuals run true to form. So this end, which developing individuals regularly achieve, is what they are “aiming at.” Thus, for a natural object, the final cause is typically identified with the formal cause. The final cause of a developing plant or animal is the form it will ultimately achieve, the form into which it grows and develops.

    References: Physics 198a25, 199a31, De Anima 415b10, Generation of Animals 715a4ff.

  3. This helps to explain why form, mover, and telos often coincide (“the last three often amount to one,” as Aristotle says (198a25)). I.e., why one and the same thing can serve as three of the causes—formal, efficient, and final.

    The telos of a (developing) tiger is just (to be) a tiger (i.e. to be an animal with the characteristics specified in the definition of a tiger). Thus, the final cause (telos) and formal cause (essence) amount to the same thing. And Aristotle also says that a source of natural change (efficient cause) is “a thing’s form, or what it is, for that is its end and what it is for” (198b3). Hence, one and the same thing serves as formal, final, and efficient cause.

    Claims like “a tiger is for the sake of a tiger” or “an apple tree is for the sake of an apple tree” sound vacuous. But the identification of formal with final causes is not vacuous. It is to say, about a developing entity, that there is something internal to it which will have the result that the outcome of the sequence of changes it is undergoing if it runs true to form will be another entity of the same kind—a tiger, or an apple tree.

  4. So form and telos coincide. What about the efficient cause? The internal factor which accounts for this cub’s growing up to be a tiger (a) has causal efficacy, and (b) was itself contributed by a tiger (i.e. the cub’s father).

    This can be more easily grasped if we realize that for Aristotle questions about causes in nature are raised about universals. Hence, the answers to these questions will also be given in terms of universals. The questions that ask for formal, final, and efficient causes, respectively, are:

    1. What kind of thing do these flesh-and-bones constitute?
    2. What has this (seed, embryo, cub) all along been developing into?
    3. What produces a tiger?

    The answer to all three questions is the same: “a tiger.” It is in this sense that these three causes coincide.

  5. Aristotle’s account of animal reproduction makes use of just these points (cf. GA I.21, II.9 and Metaph. Z.7-9):

    1. The basic idea (as in all change) is that matter takes on form. The form is contributed by the male parent (which actually does have the form), the matter by the female parent. This matter has the potentiality to be informed by precisely that form.

    2. The embryonic substance has the form potentially, and can be “called by the same name” as what produces it. (E.g., the embryonic tiger can be called a tiger, for that is what it is, potentially at least.) [But there are exceptions: the embryonic mule cannot be called by the name of its male parent, for that is a horse (1034b3).]

    3. The form does not come into existence. Rather, it must exist beforehand, and get imposed on appropriate matter. In the case of the production of artifacts, the pre-existing form may exist merely potentially. (E.g., the artist has in mind the form he will impose on the clay. Nothing has to have the form in actuality.)

    4. But in the case of natural generation, the pre-existing form must exist in actuality: “there must exist beforehand another actual substance which produces it, e.g. an animal must exist beforehand if an animal is produced” (1034b17).

  6. So the final cause of a natural substance is its form. But what is the form of such a substance like? Is form merely shape, as the word suggests? No. For natural objects—living things—form is more complex. It has to do with function.

    We can approach this point by beginning with the case of bodily organs. For example, the final cause of an eye is its function, namely, sight. That is what an eye is for. And this function, according to Aristotle, is part of the formal cause of the thing, as well. Its function tells us what it is. What it is to be an eye is to be an organ of sight. To say what a bodily organ is is to say what it does—what function it performs. And the function will be one which serves the purpose of preserving the organism or enabling it to survive and flourish in its environment.

    Since typical, non-defective, specimens of a biological species do survive and flourish, Aristotle takes it that the function of a kind of animal is to do what animals of that kind typically do, and as a result of doing which they survive, flourish, and reproduce.  Cf. Charlton (Aristotle’s Physics, p. 102):

    . . . the widest or most general kind of thing which all non-defective members of a class can do, which differentiates them from other members of the next higher genus, is their function.
  7. To say that there are ends (telê) in nature is not to say that nature has a purpose. Aristotle is not seeking some one answer to a question like “What is the purpose of nature?” Rather, he is seeking a single kind of explanation of the characteristics and behavior of natural objects. That is, plants and animals develop and reproduce in regular ways, the processes involved (even where not consciously aimed at or deliberated about) are all toward certain ends.

  8. There is much that can be said in opposition to such a view. But at least it is not ridiculous, as is sometimes supposed. In so far as functional explanation still figures in biology, there is a residue of Aristotelian teleology in biology. And it has yet to be shown that biology can get along without teleological notions. The notions of function, and what something is for, are still employed in describing at least some of nature.

Go to next lecture on  Substance, Matter, and Form.

Go to previous lecture on  Aristotle on Change

  Return to the PHIL 320 Home Page

Copyright © 2006, S. Marc Cohen