I think that if I saw a squirrel eating berries, I would explain its location in terms of self directed motivation, and not external forces. The wind didn’t blow it over to the berries, gravity didn’t cause it to roll down a hill to the berries. Waters didn’t sweep it in to the berries. It wasn’t moved there by geological processes, it wasn’t struct by something, and no one threw it over there. This doesn’t mean that it is immune to external physical forces, just that that wouldn’t be the kind of explanation one would default to to explain why it is there. They would say it wanted berries. The explanation with external physical forces would explain why a leaf is there, or a rock, not a squirrel.
But the desire is situated within a hierarchy of values (X for the sake of Y etc.), and the event is situated within causal chains of activity. The one is not necessarily consistent with the other. Bob turned on the light to scare away any burglars, but no burglars were scared away. Since the means does not necessarily procure the end, the hierarchy of values must be modeled separately from the causal activity.
While this may concern the effects of accidentals, it is not necessarily a matter of accidentals. The relations of means to ends must be judged in relation to the causation of external events, in order to be able to determine the validity of the means to ends relations. If I think throwing a hammer at the bank machine will cause it to give me a thousand dollars, this is a faulty means to end relation which needs to be accounted for.
Presumably, if the squirrel were a Davidsonian, it might say that it has a pro-attitude towards eating berries and believes that there are berries in the bush ahead. Given those attitudes, going to the bush is rational. Its desire for berries and its belief about where they are provide the reason for its action. What remains is to see how Davidson moves from this rationalisation to those mental states counting as a cause.
So we continue with the text.
For animals, lacking much in the way of discursive thought specifically, I don’t think there is much in the way of difference between wanting to and deciding to. Not much higher order thought goes into it. As for beliefs, I think that in order to include all living things, which do require some basic form of biological rationality to explain the behaviors of, we can say that “beliefs” are hardcoded, in the way of instincts and such. The avoidance of danger and pursuit of food is pretty universal, and basic internal motivation, and drive is at the center of nearly all living things, even things with no complex discursive thought, or complex conscious beliefs. A squirrel may be complicated enough to remember berries, and have an expectation that they are still there. But on a more basic level, pretty much all living things have behaviors that require internally driven explanations, or reasons. Such as, it flees from the predator because it doesn’t want to be eaten. It pursues the food because it wants to eat it.
So to Section III, and Davidson beginning the case for treating reasons as causes. We have some agreement so far that a reason for an action can be provided by a belief and a pro-attitude, the anaemic, yet irreducible rationalisation of the simple syllogism explaining that the squirrel went to the bush because it wanted berries and believed there were berries on the bush. That is enough to explain the actions of the squirrel.
A side-note is needed on the footnote; again, Anscombe and Davidson have different purposes. Anscombe is concerned with practical reasoning, reasoning concerned with deciding between alternate actions. Davidson’s approach is a part of a more general program that is further developed over the next few years, an approach to meaning dependent on interpreting utterances in terms of a first-order logic, paraphrase attitudes into a form tractable to extensional, first-order treatment. So where Anscombe’s conclusion is a normative endorsement the agent is committed to, Davidson’s is a logical construct used to specify what a primary reason is. is worth keeping in the back of one’s mind that Davidson’s account is not immediately normative, although there are normative implications we might consider once the groundwork is done.
Summarising these few paragraphs, folk agree that
- Reasons justify actions.
- Reasons are belief + pro-attitude.
- Ordinary causes don’t justify anything.
The conclusion being resisted is
- causal explanation and reason-explanation are therefore two different kinds
- Hence reasons can’t be causes
We’ve seen this in some of the comments in this thread. Davidson points out that being a justification does not appear to rule out also being a cause; it marks only the difference between a rationalisation and other casual explanations. And further:
a person can have a reason for an action, and perform the action, and yet this reason not be the reason why he did it. Central to the relation between a reason and an action it explains is the idea that the agent performed the action because he had the reason.
Italics in the original. We, as a matter of import, make use of causal language in order to explain why this reason is the one that properly explains this action.
The upshot: redescription doesn’t refute causation. 4 and 5 are not valid conclusions from 1-3.
So now you say the squirrel does not do things for a reason? Ok. Seems you changed your mind. I must have misunderstood.
If there is anything you would like to say about the text, that might help me follow your stance.
No, not a discursive reason. The contrasts used in the paper are things like crumbling bridges, plane crashes, and dissolving material in acid. So I think what I am saying is appropriate, but types of arguments for a free floating rationality of some kind to explain the behavior of living things is used often enough. Though what he is saying may be different, I just think it more broadly applies to most living things.
Importantly, he uses only rationalizations (or at least primarily), and not actions that are arrived at due to a process of reasoning, but only rationales for behaviors given after the fact. He also says he is appealing to the common sense notion of explaining behaviors in terms of reasons. So I think it applies more broadly.
It has reasons that are not discursive?
I’m not sure what that might be saying. Does that mean it does not go to the bushes because it wanted a berry - that being a discursive reason? Then why did it go to the bushes? For reasons that cannot be placed into a discourse… and what might they be? Am I right that you want to resist the idea that it was caused to go to the bushes, perhaps because you think causation will somehow inhibit the freedom of the squirrel to choose?
But we digress. I suspect that if we follow the text, many of these issues may be ironed out, and then we might have a substantive disagreement…
Again, thanks for your reply. Please read along.
By discursive, I mean a step by step explicit reasoning process that precedes the behavior. Not that it cannot be placed into a discourse. I do this because I would say that the common sense notion is broad, and I personally would use reasons to explain the behavior of living things more broadly.
1 Like
Yeah, that’s good. And that is the difference between Anscombe and Davidson in the footnote. Davidson is not setting out any supposed reasoning in the mind of a squirrel. he is providing an analysis of the events. He’s not saying that in the mind of the squirrel there was some rational sequence, he’s proffering an explanation for us to consider.
So two questions. First, does the explanation model an inferential process? Davidson says no. Second, does it commit to the states being real and causally efficacious? Let’s see how Davidson says “yes”. Then we can discuss which bits he has right.
No, I don’t think it does model an inferential process. I think that a cause is generally held to imply external direct physical interactions, whereas a reason is held to be about internal motivations. So, I personally would keep the distinction, and say that reasons can be very similar to causes, in the way of having consequents, but not the same thing.
What is the word “external” doing there? Isn’t it better to leave it out - a cause is generally held to imply direct physical interaction?
Unless you are simply stipulating that causes are external, reasons are internal, and therefore reasons are not causes - and here you will need to set out clearly what it is for something to be internal or external, and do so apart from differentiating cause and reason, to avoid question begging.
And even then, you would not be addressing the argument Davidson is proposing, which does not use this talk of the internal and external.
So here’s the rub; better to meet Davidson’s argument on it’s own ground, rather than try to meet it on your ground.
Again, this by way of appealing to the text.
I think I would just appeal to common usage. If you google the distinction, this will be given, and Davidson specifically will come up as an example of divergence from the more common usages (at least he did for me). I don’t feel like I require an argument for that. I would prefer to just say it is how they are most commonly used, and how I tend to use them. Though they are similar enough that it is easy to conflate them in some contexts, and I may myself do this from time to time.
The idiosyncratic usage relating to what is called causalism.
Common usage of what exactly? “External”? Nothing like that in my dictionary’s definition of “cause”.
I boiled the kettle because I wanted a tea looks pretty causal to me. Where’s this idiosyncratic usage? The squirrel went to the bush because it wanted a berry. We do think of rational agents being caused to do things by their choices.
But stick with the thread; there is a puzzle in what you are saying, just needing to see where to find purchase.
This is central to – though not all of – what Davidson has to deny. I started to write something about it and then realized that I’d have to bring in things we haven’t talked about yet. So consider this post a bookmark. I think the issue can be best addressed once the paper’s entire argument has been laid out.
1 Like
So what is the alternative proffered to a causal account? As described in the paragraph spanning pp 691-2, we redescribed the action so as to give it a place in a pattern of behaviour. That place in the pattern of behaviour is what counts as an explanation.
Explicitly, Davidson agrees there is merit in such an account. However he points out two things that do not follow: first, that therefore reasons are not causes.
and second, that placing a behaviour in a pattern amounts to an explanation of that behaviour.
Continuing: two things that do not follow. It does not follow that reasons are not causes. All he above could be so, and yet reasons still be causes. Again, we do use the language of causation in such cases. Here is what redescription in terms of a cause looks like:
Reasons, being beliefs and attitudes, are certainly not identical with actions; but, more important, events are often redescribed in terms of their causes. (Suppose someone was injured. We could redescribe this event ‘in terms of a cause’ by saying he was burned.)
Secondly, such actions might well be placed into a pattern and yet that not amount to an explanation. But giving the cause of some action would explain it.