How To Turn An Assailant Into Your Friend: Internalism vs Externalism

Notes and Thoughts on Bernard Williams’ Internal and External Reasons

Key Quote:  “Does believing that a particular consideration is a reason to act in a particular way provide, or indeed constitute, a motivation to act?”


One of the central debates within meta-ethics is between the various strands of naturalism, on the one side, and those of non-naturalism on the other.  Jargon, aside naturalism refers to the idea that moral values such as good, bad, right, and wrong are natural properties of the world; that is, they really exist in one way or another.  A consequence of this view is that if these values actually exist, then they are objective and are not subject to cultural or personal points of view.

Non-naturalists, on the other hand, think that moral terms cannot be expressed or reduced to natural terms.  Otherwise stated, value statements occupy a realm that is apart from natural science.  Later non-naturalists expanded the position to include the notion that knowledge of basic moral truths can be known through some kind of internal reflection which will (hopefully) constitute reasons for action.  

Within this debate there is another sub-debate between what are called internalists and externalists.  As with naturalists and non-naturalists, there are many varieties of each but for the sake of the discussion here we’ll focus a narrow set of each.  What what’s all the fuss about?

Well, in a nutshell, the internalists say that the only way to account for someone’s actions  is if they had some sort of internal motivation for it; that is, the reason for their action had to already, in some sense, be contained within them or derived as a means of satisfying a desire.  

The externalists argue that we can be motivated to moral action by–you guessed it–external reasons or facts in addition to internal reasons.  An external reason would be something like (but not necessarily) a rule like “one ought to always tell the truth” or even someone else trying to convince us to act in a certain way.

The way these two debates come together is that if moral values are part of the fabric of our world and really exist in some external objective sense, then it’s possible for these external facts to compel us to act.  But, if moral facts are somehow knowable through internal reflection only then internalists are correct and we can only be compelled to action by internal reasons.

Anyway, in very general terms, that’s what’s going on.  Now, lets get into some of the nitty gritty…Incidentally, I should say that that my own opinion is that there’s a role somewhere in this debate for cognitive science and empirical psychology.  But enough about me…

Internal Reasons

Ok, lets flesh out the internalists position regarding the relationship between desires and motivations/reasons for action.  Before we put in all the qualifications and technical language, internalism holds that every reason for action is arises out of an internal desire.  (Note: I’m going to use the words ‘reasons’ and ‘motivations’ interchangeably).  Now, lets add a little bit to this.  First we need to be explicit about what constitutes an internal motivation.  Well, this would include all your desires, dispositions of evaluation (i.e., world view), patterns of emotional reaction (i.e., personal psychological makeup), personal loyalties, and goals (short and long-term).  All of these things we call the subjective motivational set, or ‘S‘ for short.  We should also note that an individual’s desires and goals needn’t be selfish.

So far we have this massive collection of factors that can cause us to act and that can, given perfect knowledge, be retroactively pointed to as an explanation for our actions.  Lets add a little more information about S–information we can arrive at through deliberation.  It’s not uncommon for us to act and not be able to immediately point to why we acted thusly (how often do you get to use thusly?).  However, if we deliberate, (and assume we are rational) we can often explain our action in terms of something in S.

Also, sometimes deliberation works in another way.  It causes reasons for action to be added to or removed from S.  Perhaps I’m in a situation where two or more elements of S are in conflict.  Suppose, I want to both have my cake and eat it too–both desires are part of my S.

A way in which my S set will diminish is if I give up one of these desires, that is, after deliberation I decide my desire to eat the cake supersedes my desire to have it.  Or perhaps after deliberation I decide I will eat half my cake and keep the other half on my pillow for ever and ever.  The conflicting desires in S led to a new element of S– whatever ends up being the compromise.

The main point is that S is not static and can change (and not just because of deliberation).  If you think about what sorts of things you wanted when you were a child and the kinds of things you want now, it is unlikely that they are the same.

I don’t want to go into too much more detail than this but I will just add that it is consistent with internalism to say we can have elements of S that we don’t know about.   We can also falsely believe that an element is part of our S when it isn’t (i.e., we are often mistaken in identifying our desires).

As a note, this is the part of internalism that makes me uncomfortable philosophically because it’s not falsifiable (I know, I know…there are problems with falsificationism, but lets just deal with one issue at a time).   Although, this isn’t necessarily a big problem in the context we are discussing: internalism vs externalism.

Under What Conditions Might Externalism Offer a Good Account of Behaviour?

Before working up to such a situation lets see if we can describe a situation where there are clear external reasons for action, but action is better explained by internal reasons (i.e., arising from S).  A model of this type of situation would be where a person acts contrary to their well-being and the action isn’t a result of false beliefs or irrationality.

Every student encounters such situations every day when it’s time to sit down and study.  As students we know that we need to study but those cat videos on youtube are so damn funny!  The fact that we watch Maru jump in and out of boxes can be explained by there being nothing in our S that is a desire to study, hence no reasons can be generated to study.  However, there are an abundance of internal reasons (desire to feel pleasure, happiness, etc…) for watching Maru.  The desire to study is not part of our S.

One could reply that the student’s actions aren’t correctly explained by the model.  Surely, at some level the student must want to study and therefore has an internal reason to study.  But suppose the student goes into deep meditation in the mountains of Tibet for a year to search in inner reaches of his mind for this internal reason to study.  After examining his entire S, he cannot find a single desire from which he could derive a reason.

At this point we might say that he has a different kind of reason for studying, an external reason.  Suppose the student never wanted to go to skool in the first place, but was forced to do it by his draconian parents.  Both his parents are world renowned academics and expect the same future from him.  But little Johnny just wants rock ‘n roll aaaaaaaaall night, and party ev-er-y day.

His parents are all, “Johnny, you have to go to school, it’s our family tradition and you’ll have no sekurity if you don’t go to school.  If you care about the legacy of this family you will go to skool.  something something family honour.”

So, now that we know a little bit more about why Johnny was in skoo in the first place it becomes clearer that there is nothing in his particular set of desires that might give him a reason to study.  But undeniably there are reasons for Johnny to study.  It becomes apparent that any reason for Johnny to study will come from without; that is, it will be an external reason.

Alright, so we have at least one plausible picture of what it might be like for someone to have an external reason for doing something.  Now, the main issue…

Can External Reasons Alone Bring about Action?

Now, here’s the big important queshtun: could an external reason, on its own, compel someone to act.  Another way to say this is, if we were to retroactively explain someone’s actions, could we ever (correctly) explain them exclusively in terms of external reasons?

Lets go back to Johnny and figure this out.  Lets say after listening to his parents express their desperate desire for him to go to school and to carry on his family’s lofty tradition, Johnny decides to go back to skool and study.  What sort of story can we tell to explain his new course of action?

Lets begin by saying, that there are still no desires in Johnny’s S that might give him a direct reason for going to skool.  That much hasn’t changed.   But we still need something to explain his motive for going to school; that is, something in is S must have provided a reason for him to go to school.

The internalist will say that the reason for Johnny’s return to skool is because he had desires in his S that gave him a reason/motivation.  The internalist will say something to the effect that Johnny must consider his family’s traditions and legacies as desires capable of motivating action.   After reflection he realized that the best way to manifest those desires was to go to skool.  He’s not going to study because he thinks school’s important (although he’s aware of it), he’s going to study because he values his families legacy, and studying is what’s going to realize this desire.

The fact that we can point to internal motivations/reasons to explain Johnny’s change of heart doesn’t mean that external reason statements can’t motivate action.  Consider the fact that it’s common knowledge that going to university on the whole significantly ameliorates your lifetime salary (well, it used to be true anyway…).

Johnny never doubted the truth of this statement, he knew this.  He parents had been telling him his whole life.  The important take away is that even if an external reason statement is true (or thought to be true by an individual), its truth alone is not sufficient to compel action.

What the externalist has to shew is that simply by virtue of believing an external reason statement to be true (e.g., going to skool is good) that this alone somehow gives rise to motivation.  The externalist has to shew that Johnny went back to skool because going to school is good.  Not because he values family legacy (that’s an internal reason) or his father made a very convincing case (he already knew going to skool is a good thing and that would be going because his father persuaded him).

If we extend this model to the moral sphere, telling someone that something is ‘right’ (and their believing it is) isn’t on its own sufficient for the person to act in accordance with it.  Moral action will require internal reasons for there to be any, well…action.

At this point, Williams discusses further intricate arguments against externalism, but I’m really sleepy so, I’m going to pass on them for now.

A Few Final Notes

First is that the externalist has difficulty explaining the all-too-common human peculiarity of acting contrary to what one might think as objectively true (i.e., external reason statements).  The only response an externalist can give is “they’re acting irrationally”.  But plainly this is not true.  The calculating criminal who devises all sorts of plans to avoid getting caught for his crime, is not acting irrationally.  It’s rational not to want to go to jail and to avoid it–despite the ostensibly objective reason statement “stealing is baaaaaaaaaad” which the criminal probably also acknowledges as true.

Second is that when we try to convince people to do things we often encounter greater success when we give them reasons that relate to their internal desires.  Trying to convince someone that they should do something because it is objectively the right and rational thing to do is not often going to be as successful.

If you don’t believe me you can do your own applied philosophy experiment the next time you get into an argument.  To bring the other person to see the glorious correctness of your view, just keep repeating, “but my way is right and rational, so I win”.

If that doesn’t work, go to the worst area of town at night and while you’re getting robbed, explain to your assailant that what his is doing is wrong and that he is acting irrationally.  Let me know how it works!

4 thoughts on “How To Turn An Assailant Into Your Friend: Internalism vs Externalism

  1. Hey Ami,This is a great post. I just wrote a paper on this topic, so I am very into this topic. I have a few comments. (1) Ethical non-naturalists are not committed to the claim that normative propositions \”are creations of culture or expressions of attitudes,\” and it is not a \”consequence of this view is that moral values aren't objective, they depend on and are relative to a socially/historically/personally created/learned moral framework.\” Most ethical non-naturalists think that morality is objective mind-independent truths which can be discovered a priori. See Moore, Huemer, Parfit, Portmore, McMahan etc. (2) Internalists are only committed to the claim that motivation is a necessary (not sufficient) condition for a agent to have a reason and externalism necessarily denies that claim. (3) Because internalists, like Williams, include constraints that rely on an agent being fully informed in order to be motivated to x, I think they are in no better a position to handle the motivational problem than the externalists. On both accounts, an agent can have a reason to x and not be able to be motivated to x. This is the downfall of internalism. (4) I don't understand your argument against externalism and being rational. If there are external reasons (and I argue that there have to be), then any *ideally rational* agent would in fact be motivated to x whenever they have most reason to x. That is just what it means to be rational. The fact that individuals might not, as a matter of fact, be motivated to x is irrelevant, since they are not ideally rational. As such, this paragraph is misleading. You write…\”First is that the externalist has difficulty explaining the all-too-common human peculiarity of acting contrary to what one might think as objectively true (i.e., external reason statements). The only response an externalist can give is \”they're acting irrationally\”. But plainly this is not true. The calculating criminal who devises all sorts of plans to avoid getting caught for his crime, is not acting irrationally. It's rational not to want to go to jail and to avoid it–despite the ostensibly objective reason statement \”stealing is baaaaaaaaaad\” which the criminal probably also acknowledges as true.\”There are two obvious responses available to the externalist. (a) It might be rational to steal under certain conditions. An agent could be acting in accordance with her external reasons in stealing and hiding from the police (b) Avoiding jail *might* be more choice-worthy than being punished, but those are not the only two options. Sometimes, the rationally and morally obligatory option is to not steal at all. Thus, in stealing, the agent acted irrationally. Now, she might have to pick the most rational choice given that she stole, which could very well be that she ought to hide from the police. But that's not an objection to externalism at all. To lay my cards of the table, I am an (ethical) intuitionist, ethical non-naturalist, objective moral realist, moral rationalist and hardcore externalist.


  2. @Mrs. BWeasley09. Thanks for the response. I think I understand your point but I'm not sure I agree. While I agree people's desires aren't always rational, we can generally explain their actions in terms of trying to satisfy those desires. That is, their behaviour is rational in that it is directed at satisfying some desire. However, as I said earlier in the post, this line makes me uncomfortable because there's no way to falsify it. The only way (I can think of right now) to falsify it would be to show that the alternative is false; that is, external reasons don't offer a satisfying explanation of behaviour. I'm still neutral on the matter because I haven't done enough reading on it to have any sort of informed opinion. In this post I was mostly restating Williams' position in a way that I hope will be easy to follow…


  3. @Mrs.BWeasley09. There's an important distinction that you talk about, and that is the distinction between 'reasons' and 'reason' (i.e., rationality). It's actually quite a complicated issue because both of these concepts can mean different things to different people/contexts. Im not sure if there's consensus on the matter but here are a couple general ideas about these terms. 'reasons' can refer to either a explanatory or justificatory meaning. So, in the first sense a reason doesn't have to the consequence of rational thought, it's just a causal explanation. Eg. The reason I ate all 100 cookies is because I have a weakness for cookies. It doesn't mean it was rational for me to do so, but it explains the reason for my action. The second sense, justificatory, is something where the reason should provide a reason that makes sense and (possibly) could be (but not necessarily) evaluated as good or bad. Why did you steal the money? Cuz I wanted it. In this case, \”cuz I wanted it\” is an explanatory reason but not a (good) justificatory reason. Reason as in \”rationality\” is a very tricky notion. The simplest idea of it is something like choosing the most efficient way to satisfy a need or desire. But this gets complicated when you introduce moral situations. Sometimes the most efficient way for an individual to satisfy their needs/desires wont be what will be considered ethical. Is it rational to forgo your desires for others, especially if they're strangers? So, maybe we should include something about being moral in the notion of 'being rational'. And then, when you get into describing economics in terms of a 'rational decider' things get even more complicated because humans are actually very irrational when it comes to statistical reasoning. So, in these cases, do we say being rational is doing what the average person would do given full information? or what someone who has run a statistical model would do? I have no idea…


  4. correction: in the last paragraph, I shouldn't have said \”humans are actually very irrational…\” . I simply should have said \”humans are actually very poor at statistical reasoning\”


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s