In the opening passage of the Nicomachean Ethics Aristotle seeks to pick out the specific subject of his study for the remainder of the book. His discussion is often misunderstood, but a good understanding of it will serve us well in understanding the study of ethics. We will consider the passage bit by bit with comments and clarifications as we go along, doing our best to read it according to the principle of charity.
The good has the nature of an end
Every art and every inquiry, and similarly every action and choice, is thought to aim at some good; and for this reason the good has rightly been declared to be that at which all things aim. (emphasis added)
Contrary to what some people think, Aristotle is not committing a quantifier shift fallacy here. Rather, he’s picking out some determinable, the good, which is common to all things done for some end. Let’s unpack this.
In general, something is less determinate (and therefore more indeterminate) if it is vaguer or less specific. So, for instance, red is less determinate that scarlet. Furthermore, determinateness comes in degrees: red is less determinate than scarlet, and coloured is less determinate than red. We use the term “determinable” to refer to some partially indeterminate feature which can be determined in some way. So, coloured is a determinable which red determines and red is a determinable which scarlet determines.
When two things resemble one another it is on account of them sharing some determinable feature which they each determine in some way: a scarlet thing resembles a crimson thing in that they are both red (that is, they share the determinable red), and both resemble a green thing in that they are all coloured things (that is, they share the determinable coloured). Just as determination comes in degrees, so too does resemblance: the scarlet and crimson things resemble each other at more levels of determination that the scarlet and green things. Speaking discretely, the scarlet and crimson things resemble each other as red and as coloured, whereas the scarlet and green thing only resemble each other as coloured.
In this opening passage Aristotle seeks to narrow the focus of his study by picking out the determinable that all desired things share, according to which they resemble each other as desired or as aimed at in some activity. He notes that whenever we desire or aim at something it is because of some good in it, and therefore the good is rightly declared to be this determinable he’s looking for. Now, just as what makes something one colour as opposed to another will depend on the particular way in which the determinable coloured has been determined, so too the reason why this or that thing is desirable or aimed at will depend on the particular thing in view. Good ice-cream and good vacations are desirable for different reasons, and so determinethe good in different ways, but they resemble each other in that they are pursued.
Later treatments would make explicit a question which, as far as I can tell, Aristotle leaves implicit or thinks obvious: is something good because I desire it, or do I desire it because it’s good? It cannot be the former, since I often desire things I later realise were in fact bad for me.
For haven’t we all had the experience of wanting something which we ourselves then admitted was not good? I wanted that last drink at the party, but afterwards I admit that it was not good for me. I wanted to drive 100 mph down the winding road, but later, on my hospital bed, I admit that it was not good. If wanting something made it good, then my wanting the last drink would have made it good for me. (Edmund Waldstein, The Good, the Highest Good, and the Common Good, thesis 2)
It must be, therefore, that I desire something because it seems good to me in some way. That is, the goodness I perceive causes the desire in me. Of course this perception may be incorrect, but the point remains that it because of the good I perceive in something (correctly or incorrectly) that attracts me to it as something worth pursuing.
Returning to our passage, Aristotle is noting here that in general the good “has the nature of an end” (cf. ST I-II Q9 A1 corp). An end is “that for the sake of which something is done” and a means is “that which is done for the sake of something”. These are complementary notions such that whenever we have one we also have the other. We see Aristotle make this same connection in the Physics where he lays out his four kinds of causes. In the passage he identifies ends as what later would be called final causes:
Then there are things which are causes in the sense that they are the ends of the other things, and are the good for which they are done. Without quibbling about whether it is an actual good or an apparent good, that at which other things are aimed — that is, their end — tends to be what is best. (Aristotle, Physics II.3 195a23-25)
Note that the Aristotelian “cause” is much broader than the modern’s “cause”. The modern usage most closely approximates the Aristotelian efficient cause. For those unfamiliar with the Aristotelian usage, perhaps “four kinds of explanations” is more helpful for conveying what he’s getting at.
Note also that the phrase “tends to be what is best” is just there to explain how ends and goods relate: something is good to the extent that it fulfills its end, and so to achieve its end in full is best (that is, most good). This is all he means.[1]
In summary, then, this first passage involves distilling this determinable the good, which is what accounts for the resemblance between things as desired in some activity. It picks out something as an end or that for the sake of which the activity is done.
Ends are better than means
But a certain difference is found among ends; some are activities, others are products apart from the activities that produce them. Where there are ends apart from the actions, it is the nature of the product to be better than the activities.
Now, as there are many actions, arts, and sciences, their ends also are many; the end of the medical art is health, that of shipbuilding a vessel, that of strategy victory, that of economics wealth. But where such arts fall under a single capacity — as bridle-making and the other arts concerned with the equipment of horses fall under the art of riding, and this and every military action under strategy, in the same way other arts fall under yet others — in all of these the ends of master arts are to be preferred to all the subordinate ends; for it is for the sake of the former that the latter are pursued. It makes no difference whether the activities themselves are then ends of the actions, or something else apart from the activities, as in the case of the sciences just mentioned.
Aristotle here makes distinctions regarding how activities and ends relate to one another. First, either the end and activity are the same or they are distinct. An orchestra playing a piece is an example of the former, since the performance is both the end and the activity. A carpenter making a chair is an example of the latter, since there’s a real distinction between the production of the chair (activity) and the chair (end). We must note that by “product” we don’t only mean physical objects that result from some activity, as the chair results from the carpentry. Rather, we mean any outcome which is distinct from the activity that brings it about. So, winning a sports game is the end and product pursued when playing the game.
Second, an activity can be made up of other activities. In this case, we might say that the subsuming activity is superordinate (or “master”), and the subsumed activities are subordinate. Subordinate activities are parts of superordinate activities.
Both distinctions show us different ways in which ends and means might arise and relate. Sometimes the end and the means are really the same thing, as when an activity is the end we desire. In this case, the distinction we impose is merely conceptual. Other times they are really distinct, as when an activity produces something external to it. Moreover, when one activity is subordinate to another the former is done for the sake of the latter, and so the former relates to the latter as a means to an end.
Twice in this passage he picks cases where there is a real distinction between ends and means (the product and activity, and the superordinate and subordinate activities), and notes that the end is always better than the means. This is true because “it is for the sake of the former that the latter are pursued”. That is, the end is more truly the thing desired, whereas the means is desired only in a derivative way. The end is desired through the means.
To make this more precise we need distinguish between the thing itself on the one hand and the thing desired on the other. Now, a thing is desired to the extent that it — and not something else — fulfills that desire, and so the desire for the thing itself is proportional to how closely it relates to the thing desired.[2] Thus, if just this or that feature of the thing is desired, then that feature is more desired that the thing itself. For instance, if I buy a torch because I want the lightbulb inside of it, then I desire the lightbulb more than I desire the torch. Conversely, if the thing itself is just one feature of what is desired, then the greater whole will be more desired than the thing itself. For instance, if I desire a violin performance because I desire an orchestra performance, then I desire the orchestra performance more than the violin performance.
Applying this to the cases Aristotle mentions, we can see why his claims are true. First, there’s the case when an activity is desired for the sake of some product really distinct from it. Here the activity is desired because of one of its features, namely the ability to bring about the desired product, and so the acquisition of the product is desired more than the activity itself. Second, there’s case of a subordinate activity being desired for the sake of some superordinate activity. Here the subordinate activity is desired because it is part of the superordinate activity, and so the superordinate activity desired more than the subordinate one.
We might also arrive at this conclusion from a slightly different angle. We’ve seen that goodness has the nature of an end. Thus to be better (that is, more good) is to be more of an end. Now, something is more of an end if it is closer to some final or ultimate end, as riding is closer to strategy than bridle-making.[3] But if A is a means to B, then B is closer to some final end, and is therefore more of an end, and therefore better.
At this point we must make two clarifications.
First, people sometimes mistakenly interpret Aristotle as assuming that bridle-making is only ever done for the sake of strategy. The passage does not require that we interpret him this way, and given his historical context he surely knew that bridle-making could also be done for the sake of other things, like recreation or sport. What he’s doing here is picking one of these instances as a concrete example of how activities might relate to one another such that some subsume others. If you prefer you could use an example where bridle-making is subsumed under some activity other than strategy, but the point would remain the same.
Second, to say that that ends are always better than means is not to say that things that are ends are always better than things that are means. Rather, we’re claiming that things considered as ends are always better than things considered as means. Part of the import of the first passage is that whenever we consider something better than another thing, it must be with respect to some end. But the complexity of human desires means that the same activity might be desirable for more than one reason, and therefore on account of more than one end. Imagine, as an example, that our friends have come together to study as a group. On the one hand, this might be desirable because studying produces knowledge. On the other, we might desire it because we enjoy spending time with our friends. The conclusion here, and Aristotle’s point, is that knowledge is better than studying considered as a means to knowledge. We’re saying nothing about the relationship between knowledge and studying considered as a part of spending time with friends.
In general, the claim that ends are always better than means is not the same as the claim that if B is ever a means to A, then A is always better than B. Rather, it is that claim that whenever and insofar as B is a means to A, A is better than it.
In summary, then, this passage notes certain helpful distinctions regarding activities and ends. Sometimes the activity and the end are the same, and sometimes they are distinct. In the latter case, the end is better than the activity. Activities themselves can often be divided into sub-activities (called subordinate activities), and in these cases the superordinate activities are better than the subordinate activities.
The chief good has the nature of a last end
If, then, there is some end of the things we do, which we desire for its own sake (everything else being desired for the sake of this), and if we do not choose everything for the sake of something else (for at that rate the process would go on to infinity, so that our desire would be empty and vain), clearly this must be the good and the chief good.
This passage has caused much confusion for readers, and it certainly would have been better if Aristotle had spent some more space clarifying his meaning here. Some have thought that by “chief good” Aristotle is picking out some particular final end of all human life. I’m inclined to think that he only introduces such a notion later in the first book, and even then with more nuance than some commentators would grant him. Alas, we will have to leave that for a future post.
But if he is not talking about a specific end, then what could he be talking about? Good question. In the first passage Aristotle arrived at this determinable the good which, having the nature of an end, is that for the sake of which everything is desired. Now, as with all determinables, we can determine this in various ways to various levels of specificity. So we can talk about the good, the good thing, the good artist, the good musician, the good violinist, the good first violinist, and so on. This parallels how we can talk about the coloured thing, the red thing, the scarlet thing, and so on. But notice how we could determine things differently, so that instead of determining coloured to red we could determine coloured to brightly coloured. This kind of “alternative determination” is what Aristotle is doing here in this third passage.[4] We’ll first discuss the structure, and then explain his defense.
The first part of the sentence reintroduces the notion already discussed in the first passage — this determinable the good — which he refers to as the “end of the things we do, which we desire for its own sake”. Now in the second passage you’ll recall he discussed ends and means and how they arise in various general ways, and he noted that the ends are always better than their means. Now consider some particular case where A is desired for the sake of B, B is desired for the sake of C, and so on, but where this chain comes to some final end Z. In this case Z is different from all the other members in the chain in that it is not desired for the sake of something else, or in other words it does not derive its desirability from another as a means derives its desirability from its end. This is the property Aristotle wishes to use in his alternative determination of the good, and it gives us this determinable the chief good. Now the chief good is still fairly indeterminate, and there is nothing in the notion itself that requires that it pick out one particular good in all cases. Depending on how we determine it we will get different goods: the chief medical good is health, the chief economic good is wealth, and the chief military good is victory. The important thing here is that the chief military good is not bridle-making, since the latter is an activity subordinated under the activity of strategy and as such derives its desirability from that superordinate activity.
So, if we were to repeat the passage, taking out the parentheses and highlighting corresponding determinables and their names, we would get the following:
If, then, there is some end of the things we do, which we desire for its own sake… and if we do not choose everything for the sake of something else… clearly this must be the good andthe chief good.
This, then, is the structure of the passage. But you’ll notice that Aristotle thinks that all chains of desire must end in some or other chief end. He summarises the reason for this in the second pair of parentheses when he says that if a specific chain didn’t come to an end then “the process would go on to infinity, so that our desire would be empty and vain”. Someone unfamiliar with the distinctions and arguments introduced by him in his other works like the Physics and Metaphysics can be forgiven for missing that he is just summarising and applying these here, as opposed to working them out from the start again. For the sake of clarity we will expand his summary slightly.
In those works Aristotle makes use of a general distinction between what would later become called per se causal chains and per accidens causal chains. These days they are also sometimes called essentially ordered causal chains and accidentally ordered causal chains respectively. The defining characteristic of a per se causal chain is that each member in the chain acts only insofar as it is acted upon, so that it derives its power to act from some other member in the chain. The standard example of such a chain is that of a stick which pushes a rock, which it does only insofar as it is pushed by me. On the other hand members in a per accidens causal chain do not depend on each other in this way. Here the standard example is that while my father depends on my grandfather for his coming to be, it is not the case that my father begets me only insofar as my grandfather begets him. (As a reminder note that while I’m using efficient causal chains as illustrative examples, the term “cause” here is used in the broader Aristotelian sense and not in the limited modern sense.)
Now, any particular per se causal chain requires an ultimate cause, by which we mean something with underived causal power in the relevant sense. This ultimate cause is also sometimes called a “first” cause, but when using this term we must remember that we aren’t concerned with something first in the sense of being earlier than all the other cause, but rather something being independent of the other causes and on which they depend. Indeed, when considering chains of final causes, this “first” cause is actually the last end.[5]
The reason why per se chains need ultimate causes is because each intermediate cause merely propagates the causal power it derives from another, so that unless there’s some originating cause there would be no causal power to propagate in the first place. Now because people are prone to misunderstand what’s being said we note that this point isn’t primarily concerned with the number of causes, but their kind — namely, that they are derivative causes. For example, it doesn’t matter how many water pipes you have, they will never by themselves be able to direct a flow of water unless something puts water into the system. Similarly, if everything in a collection can push only insofar as it is itself pushed, then that collection cannot by itself push anything.
Aristotle is here applying this to the notion of final causes: if everything in a collection produces desire in me only insofar as it derives that desirability from another, then that collection cannot by itself produce desire in me. What we need is something which is desired for its own sake and not for the sake of another, failing which the chain would have no power to produce desire in me (or, as Aristotle says, “our desire would be empty and vain”). This ultimate final cause, or ultimate end, would then be an example of a chief good.
So far we have left one thing in the passage unexplained: if the point about per se chains isn’t primarily about the number of causes, then why is Aristotle concerned that “the process would go on to infinity“? We can take Aristotle’s words in two ways, each of which complements the other. First, it might be that he’s using the term to pick out the notion of an infinite regress in the sense that there is no ultimate cause. This is sometimes how the term is used these days, and in this sense of the term it is consistent with there being infinitely many intermediate causes between the ultimate cause and the final effect (assuming such a thing is coherent). Second, while the point about per se chains isn’t primarily about quantity, it has a secondary consequence about quantity. It follows that the chain must be finite from the facts that (1) for each cause there is a next member of the chain (that which it causes), (2) there is a first member (the ultimate cause), and (3) there is a last member (the final effect).[6]
In summary, then, this last passage combines the insights from the first two and, by way of “alternative determination”, picks out the primary focus of the rest of the rest of his study, namely this determinable the chief good. This is just the beginning, however, and in a later post we will discuss the narrowing of his focus to the particularly human chief good that occurs later in the book.
Related resources
The biggest influence on the approach I followed here was David Oderberg, particularly his papers On an Alleged Fallacy in Aristotle and The Content and Structure of the Good. I quoted Edmund Waldstein’s The Good, the Highest Good, and the Common Good, and I highly recommend reading that too.
The notion of a determinable and it’s distinction from a cause is critically important for precision of thought. Ronald McArthur’s paper Universal in praedicando, universal in causando is an invaluable resource for understanding the distinction between predication (which corresponds to our determinables) and causation. I highly recommend it.
For more information on per se vs per accidens causal chains I recommend Edward Feser’s blogposts Cross on Scotus on causal series and Edwards on infinite causal seriesm Caleb Cohoe’s paper There Must Be a First, and Gaven Kerr’s paper Essentially Ordered Series Reconsidered. Cohoe’s paper is the one where I realised that additional reasons need to be given for thinking that per se causal chains are finite.
Notes
- In general something fails to fulfill its end only to the extent that it is prevented in some way, due to either internal defect or external interference. For instance, a carpenter might aim at making a chair which can hold people up and not fall down easily, and will only fail to achieve this if impeded by something internal (like lack of ability) or something external (like bad materials). Unless such interference occurs the carpenter will achieve their end in full, which is the best result. We’ve mentioned before that classical thinkers like Aristotle realised that there is a broad sense in which all things are orientated to certain ends given by their natures. And in the Physics he has this more general notion in mind. For instance, the development process of a dog is directed toward the growth of four legs with which the dog can walk, and only fails to achieve this when the dog has some kind of genetic defect or has some external blocker is present, like an accident or lack of food. Again, unless it is interfered with the process will achieve it’s end in full, which is the best result.
- In coming up with this phrasing I thought of a number of alternatives, which I include here for posterity: “the thing exhausts the desire without excess”, “the thing itself, and not some part or some greater whole, is what’s desired”, “the thing itself is desired, and not some part thereof or some whole of which it is part”, “the thing is desired neither merely in part or as a part”, “the whole of the thing is the whole of what’s desired”, “the thing is all and only what is desired”, “the thing in reality matches the thing desired”, “a thing is desired to the extent that it matches the object of that desire”, “the thing satisfies the desire without excess or deficit”, “of the thing itself and the thing desired, neither is a part of the other, but they agree completely”, “the thing itself, and not something more or less, fulfills the desire”.
- Shortly we will defend the claim that every chain of ends must have some final, ultimate, or “chief” end. However, this is not required for this point. If we have some infinite series of ends, then to say that B is more of an end than A it is sufficient that it be closer to some end C which is further along the chain.
- The rough idea is as follows: when we determine some determinable we contract or qualify it in some way. Now we come to know determinables through abstracting away these qualifications, and so the most “natural” way of determining them is by the same road we took to get there in the first place. But nothing in this constrains us to this, and we are free to qualify or contract in a different way to how we first arrived at the determinable. This different way is what I’m calling an “alternative determinations”.
- I’m reminded of the saying of Christ that “the last will be first, and the first last.” (Matt 20:16)
- While Aristotle wouldn’t have expressed it in these terms exactly, it seems from his other works that he understood the principles at play here. The three facts combined mean that the causal chain is one-to-one mappable onto a bounded contiguous range of natural numbers, which is only possible if the chain is finite.
Thinking Thought Out 