A common idea in Empiricist schools of thought is that ethical questions can never be based on factual data. Consider Hume's "is-ought" problem or the Logical Positivist notion that ethical statements were non-sensical. It seems to me that this could be easily addressed by the following approach:

  • Map various emotions and value based statements to the neural configuration that corresponds to them: Pain corresponds to this neural state, Pleasure to this neural state, "Killing is bad" because the fact of death causes this neural state in those people affected by the death, etc...
  • Construct a utilitarian ethics based on maximizing positive neural configurations and minimizing negative ones. One doesn't need an explicit mapping of emotions and value statements, as long as one has established in principle that they correspond to neural states.

My questions:

  1. Are there any hidden assumptions in this approach that I am missing? Is this approach feasible only if we admit a physicalist approach to the mind-body problem? Or can the values-neural states correspondence still hold even for a mind-body dualist?
  2. Does the fact that even the very basic positive vs negative dichotomy is relative constitute an obstacle to such an approach? or can one posit/axiom their way out of this dilemma?
  3. Has anyone of note suggested such an approach?

Four assumptions:

  • You assume such a mapping is possible to construct using empirical evidence. (it's not known if the human brain is quiescent enough to permit mapping)
  • You assume the resulting map will be tractable, allowing for predictions to be made using it (we've mapped the genome... it's still taking decades to figure out that spaghetti code, Modeling brain behaviors is currently a herculean feat for supercomputers)
  • You assume it is possible to construct a metric with which to construct the utilitarian ethics system from the map. (Get a room full of people to agree on a pizza topping, and we'll talk about trying to get them agree to a procedural way to determine ethics. Along the way, we might even get them to agree on a presidential candidate!)
  • You assume that the act of executing these steps is ethical. (this one is frustratingly hard to sidestep without introducing all sorts of side effects)

Also, consider that, if you believe that there is no metaphysical source of ethics nor morality, the human species has been undergoing exactly the process you describe for a hundred thousand years, and our system ethics is actually the result of that process. If this is indeed the case, you will have to explain why your process should be preferable to the existing one. Doing so will likely involve understanding the existing one, which has been the subject of philosophy for thousands of years.


Your proposal comprises at least the following steps:

  1. Select a set of actions considered relevant under an ethical aspect, e.g., killing.

  2. Determine how the person, who is affected, experiences these actions, e.g. as pain or pleasure.

  3. Determine the neuronal correlate of these emotions.

  4. Evaluate the neuronal correlate as positive or negative.

You can skip point 3 and the detour via neuroscience. A utilitarian ethics does not work better with the evaluation according to point 3 than by a direct evaluation according to point 2. Hence I do not see any benefit from bringing neuroscience into discussion.

The philosophical problem remains how to derive prescriptions, i.e. normative statements, from descriptive statements without bringing in at least one normative axiom. Therefore I do not see how your proposal contributes to link empiricism and ethics.

As far as I know your proposal has not been advanced in the literature.


I do not believe such a mapping could ever get us very far, nor unify Hume's "is" and "ought" dichotomy.

The technical problem is that a fact or datum, the "is", is something that has happened, is measured, recorded, and accounted for. The "ought" by definition is a judgment about the future. It is also an "ethical judgement," thus one that assumes a certain freedom of choice, an absence of causal determinism, or as Kant puts it, a different type of "causality" from those described in natural sciences.

In his attempt to resolve such issues, Kant derives the categorical imperative from the very assumptions of freedom and reason, wherein "ought" is given by reason and only implies "can." This is was described by many as an empty formalism, inadequate to specific real cases. Your "neural" imperative seems equally empty. We don't need neural mappings to tell us that pain is bad or that we should save children from a burning building.The types of cases most plausibly mapped.

Yet it is unlikely that gazing into the neural tea leaves we will find the answer to such questions as: Should I defy Creon and bury my brother according to custom, even though he broke the laws and subverted the state? Unless such a problem repeatedly came up. Though one can reason towards ethical solutions, such reasoning entails, in serious cases, judgments concerning freedom, indeterminacy, and an open future. Which is why such judgments are rendered not by machines but by judges and juries capable of ordinary imagination, guesswork, and empathy.

To base ethical judgments in some appeal to "science" would only make them less subtly reasoned. It would be to willfully return to the crude imperatives of "nature" by the paradoxical route of discovering its "laws." One could perhaps rig up an "ethics machine," a neural-cognitive Leviathan that directs us repeatedly towards certain states. But I think the outcome would be a monstrously "unethical" apparatus and dystopian nightmare.

So I would suggest that the "hidden assumption" you are missing is the complex and necessary relationship between reason, ethical judgment, and freedom, perhaps best developed in its relation to modern science in Kant's critical philosophy. You cannot adequately correlate neural states and ethical judgment without abandoning the assumption of "rational freedom" on which the latter rests.


Regarding #3....

Neuro-ethics is an active area of research that has yielded insights into both the structure of the brain and ethical dilemmas.

Some notable names...

  • While I see the point to some extent, I guess I would want to add that the "insights" are generally reached by people who were already committed to the belief that ethics should really be neuro-ethics. This doesn't invalidate them, but not everyone working in philosophy thinks the field has been all that productive. – virmaior Nov 10 '15 at 0:12

The standard objection against positions like your suggested is that this is a "categorical fallacy":

While you can easily describe human behaviour (like ethical judgments) as based on material, measurable events (i.e. is in some sense dependend), it is simply not valid to identify them.

The hidden assumption is a scientific realism: Only what is scientifically measurable is real and vice versa.

The problem with this position is the same as with idealism: It is reductionalistic. Natural sciences can only describe material events. They describe things insofar they are materia/energy, and in this, they are the best possible procedure. What neurologists often forget is that they are not simply describing the world as it is, but interpret it.

And that is the core of the problem: The identification of occurences in the brain and human behaviour is itself interpretation, that is human behaviour, which can (by assumption, which is human behaviour) be identified with neural activity by interpretation and so on.

Helmuth Plessner in his not yet translated work (according to Wiki) "Die Stufen des Organischen und der Mensch" therefore stated that the only epistomological position that is able to overcome the Mind - Nature dualism (and the implicit reductions of realism and idealism) is a phenomenological analysis of life, that is behaviour in the broadest sense, both of subject and object. A similar approach has been attempted by pragmatists like Dewey.


Even if the mental supervenes on the physical, this does not imply an identity - as per Klockings answer.

But say that it did, for the sake of argument, then the difficulty of the ethical problem at hand is not made any easier, but considerably more difficult ie is the mind-reading machine - which in essence you are hypothesising - is this accurate, or biased; what happens if it reports something I'm supposed to be thinking but I categorically deny I ever did?

  • 1
    My idea wasn't about trying to solve any ethics problems per-ser. I realize that it is functionally identical to utilitarianism. I just wondered if it might be a way of dismissing the empiricist claim that ethics have no factual basis. – Alexander S King Nov 14 '15 at 6:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.