Sunday, March 18, 2007

Foundational Moral Principles

The impetus for this pulse is a comment to my post two days ago, “Patricia Churchland: The Biology of Morality” .

That comment hinted to me that the author was assuming that there must be some type of fundamental moral principle, and that my job was to account for them.

However, it is my position that there is no such thing as a ‘fundamental moral principle’, and nothing to account for.


In Churchland’s presentation, she suggested that oxytocin levels, for example, may be used to determine if an individual should or should not be promiscuous.

I answered that if we accept Churchland’s initial assumptions, and we can affect oxytocin levels through social actions such as praise and condemnation, it is reasonable to ask whether we have reason to do so, and whether they are stronger than any reasons we have not to do so. If, for example, praise will increase oxytocin levels, and oxytocin levels make others kinder and more considerate of others, then we have reason to use praise. I further claimed that moral questions operate on this level, not at the level where Churchland placed them.

Michael, in his comment, suggested that we can raise the same question about my “kinder and more considerate of others” are moral reasons facing the same problem as Churchland’s – that they, too, are in need of justification.

I am not certain if this is what Michael had in mind. However, the context of his comment suggested that we have two options. Either we have an infinite regress of moral reasons – that is, we justify one moral principle in terms of a more basic moral principle, which we justify in terms of a more basic moral principle, and so on to infinity. Or, we must reach an end point – a basic, fundamental moral principle that needs no further justification.

If we generalize this objection we find its use all over the place. With respect to the history of the universe, we find those who argue that the universe must either be infinite, or it must have a ‘first cause’. In the area of knowledge itself, people assert that justification or ‘proof’ must either be infinite, or go back to a set of foundational premises that, themselves, must be taken on faith and can never be justified.

Coherence of Beliefs and Harmony of Desires

In some cases, philosophers have held that this dilemma between an infinite regress and self-justifying foundational propositions is a false dilemma. There are other options.

Philosophers interested in theories of knowledge have another option for justifying beliefs – coherentism. To picture this theory at work, think of a web. A belief sits at each node in the web, and is connected to other beliefs through strands of logic. Each belief is justified by its relationship to other beliefs, themselves justified in terms of connection to other beliefs, which (eventually) which are all tied together. There are no foundational beliefs – no self-justified propositions that serve as a “foundation” for all others. Everything is justified by the quality of its connections to everything else. Some beliefs are very well connected. Others are connected only by thin threads.

The desire-utilitarian view of value is the same. All value exists in the form of relationships between states of affairs and desires. Desires themselves are evaluated by their relationship to other desires, which in turn are also evaluated by their connection to other desires, and so on.

In the case of beliefs, the relationship for the web with the most and strongest connections is “coherence”. Under coherence theory, if one person believed that P, and another believed that not-P, this would be taken as a sign of incoherence. One of the two would have to be mistaken.

Desires have a place for a similar type of ‘coherence’, but it also has important differences. I can illustrate these differences with an example. When it comes to eating chicken, I have a preference for dark meat. My wife likes white meat. As a result, we never get into a disagreement about who gets which pieces of chicken. This situation allows both of our desires to be more easily fulfilled – precisely because we like different things. I call these types of relationship ‘harmony,’ to distinguish it from ‘coherence’.

With respect to beliefs, we are looking for the greatest coherence. With respect to desires, we are looking for the greatest harmony.

If we accept this model of value, we do not need to postulate anything more bizarre or strange than desires, states of affairs, and the relationships between them (whether a desire is fulfilled or thwarted in any given state of affairs). The idea of evaluating desires simply applies this models to “states of affairs” in which individuals have particular desires. Just as we can evaluate everything else (knives, movies, friends, jobs, houses, meals) by their ability to fulfill desires, we can evaluate desires by their ability to fulfill (other) desires.

There is no dark mystery here, no special powers that transcend the real world but interact with it giving things value, no ‘foundational ought’ that seem to emerge from nowhere and cannot be bound to anything in the world of ‘is’.

Nor do I need to postulate, explain, or account for some sort of foundational ‘oughts’.

Accounting for Foundational Oughts

If you think about it, a great deal of effort is spent in debate over the nature of these foundational ‘oughts’. We have people who are called ‘realists’ and ‘objectivists’ who assert that these foundational ‘oughts’ really are real-world entities. They exist. They must exist. Yet, these ‘realists’ cannot even start to answer the question of what they are, what they are made of, how they work, or how we can know about them. If they exist, they are extremely strange entities. They are not strange in the way that black holes are strange. They are strange in the way that ghosts are strange – a type of strangeness that suggests they do not exist.

On the other hand we have the ‘subjectivists’ who state that the brain can simply make up these foundational oughts. They have no independent justification, and can be whatever the agent or the assessor wants them to be. Strangely, these foundational ‘oughts’ are a lot like hallucinations. They appear only to the person who claims to be able to see them. They guide his actions. The agent takes them to be real even though they have no existence outside of his ability to perceive them. Yet, even though they have all of the qualities of subjective moral hallucinations, people can draw upon them to justify killing, imprisoning, or otherwise harming those that these hallucinations point to and call ‘guilty’.

On the ‘harmony’ model that I am defending here, I have no need to trace a line of moral reasoning back to some set of self-justifying moral ‘oughts’. In fact, I assert that they do not exist. The reason that the ‘realist/objectivist’ and the ‘subjectivist’ above have so much trouble coming up with a theory of these foundational moral ‘oughts’ that make sense is because they are both starting from a false assumption – that there must be foundational moral ‘oughts’ and we must have a theory that makes sense of them.

The most important conclusion to draw from this is that when somebody else comes to me and asks, “How do you account for foundational ‘oughts’?” I answer that their question is misplaced. My situation with respect to foundational ‘ought’ is much like the situation that 19th century astronomer Pierre-Simon Laplace when Nepoleon asked him where God fit into his theory of celestial mechanics. Laplace’s answer, “Sir, I have no need for that hypothesis,” is the same answer I give to any question of where I put foundational ‘oughts’.

Reasons and the Web of Harmonious Desires

For the sake of completeness, I want to add one more item to this account – reasons for action.

I have stated that desires are the only end-reasons for action (reasons for action that identify ends or goals) that exist. If an agent has a desire that P, and P is true in state of affairs S, then agent has an end-reason for action for bringing about or maintaining S.

In an agent’s quest to establish or maintain a state of affairs S in which P is true (given that agent has a desire that P), the agent has end-reason to ask what he can do to influence the desires of others. This desire that P that is his y end-reason for action gives him an end-reason to perform those acts that will promote desires in others that will help establish or maintain S. This end desire also gives the agent an end-reason to perform those acts that will make less common in others those desires that will tend to prevent or destroy S.

In this web of all desires, the more a desire comes into conflict (destroys harmony with) other desires, the more and the stronger the reasons that exist to act so as to inhibit that desire. At the same time, the more a desire tends to fulfill other desires, the more and the stronger the end-reasons are for promoting that desire.

There is no mystery as to why a person with a desire that P has an end-reason to influence the desires of others in this way. It is a part of the desire that P, where desires are the only reasons for action that exist, that it gives him an end-reason to seek harmonious desires in others.


D.R.M. said...

I haven't read all this post, but I see you've adopted Coherentism as a theory of justification.

I'd just like to know what you're "theory of truth" is. I know this, in no way, relates to ethics, even meta-ethics. But, nevertheless, I'd find such a comment interesting.

Iain Quicksilver said...

I've recently seen some internet articles claiming that there can be no morality without a God. If "God is good" has any meaning, he must follow ethical principles, in which case those principles stand above him and he is not the highest. If he determines the content of morality, then the statement "God is good" becomes a meaningless tautology.