Neuroscience is to Ethics…

by Sanjay Manohar, 2002

Neuroscience is to ethics as physics is to architecture. The architect's creativity is moulded by his experiences of structures and their engineering. His aesthetics transform the physical principles and human needs into a work of art. Engineering requires the mathematical precision of geometry.

Ethicists are. There is no surprise that despite much effort to systematically justify them, moral sciences remain as arts subjects. Those who define ethical principles have intuition about what can or cannot be said. However, the ultimate cause of any boundary conditions must come from human psychology.

Against Duty

Duties are things that could affect our actions. They are seen to be external to our own will. However I argue that they should be viewed as being as internal as our own desires, but acting on a different neuropsychological domain.

Duties are reasons for action

We look for reasons for our actions, and we call them thoughts. People often believe in a sequence of events: perception leading to thought, causing an action.

The thoughts that we find by introspection may or may not reflect the 'reason for action' that a psychologist may infer (the phenomenon of cognitive dissonance, Nisbett and Wilson 1977).

The difficulty in attributing a cause to the visible effect is three-fold.

1) Consciousness only gives a subject the tip of the iceberg of the brain's activity. He is partially informed.

2) The observer's viewpoint is purely phenomenological. Although we find rules and pragmatic summaries of action, causation can only be inferred indirectly. Compare with the way Newton formulated a law of gravity which economically predicts many things from apples to planets. It is an explanation only by virtue of its simplicity and power. Psychological techniques treat the brain like this, as a black-box.

The concurrence of the two accounts, that of the subject and of the observer, is our primary goal of explanation.

Individuals experience pulls and pushes that sway them to make a final decision, like a continuous physical system with a finite number of stable states. We can talk about these factors in many ways:

  1. Conscious/unconscious
  2. Strong/weak
  3. Directional specificity
  4. Brain location/neural properties
  5. Timescale

Logical thought has a role in construing these factors, but what determines their weight or direction towards action? We have many reasons for doing any one thing rather than another, but who adds up the reasons and sets the thresholds for action?

The energy landscape viewpoint

If we have a dynamic system that converges into one of several stable states by traversing an energy landscape in a multidimensional space of states, we can express the system's state as a vector whose bases represent units of proximity to the possible stable states. We can express the brain's thoughts either as millions of neural events, or as its tendency towards certain outcomes or actions. How can the products of thought ever be thought of as stable states? Evidently, no thought is final, and every thought leads to another. Consider a single moment and its brain state. The next thought depends on the contemporaneous input, and in the absence of input there is a certain 'default' sequence or progression, which can be thought of as more stable. Any exogenous input raises the energy of the system allowing divergence from the default path.

Is logic an illusion?

What are illusions?

There are many things that seem different when we look in two different ways. An illusion is a special case of this. Illusions occur only when we can definitely say that some viewpoints give a more reliable prediction of how something will seem from other possible viewpoints, as compared with the 'illusory' viewpoint. In fact, the illusion is reliant on our predictive mechanisms - that is, how a single perception gives rise to expectations of what we might perceive in the future.


The brain creates systems of belief, or cognitive frames, which is jargon for a small amount of information to represent a large amount of input.

There are many ways of choosing this small 'internal' information to match what we receive. We could liken it to the way reporters listen to a rambling scientist and concisely extract the features they require.

The 'compressed' data, if you like, constitutes a model, in that certain features have been extracted in a way that will permit useful predictions to be made.

In this sense of the word 'model', we have an internal model of the visual world consisting of segregated, labelled objects; or a model of someone's speech as an ordered series of tokens and their corresponding 'meaning'.

A model is anything that represents and can be used somehow for output (action).

As an example, optical illusions lead to an 'incorrect' mental model of visual reality. Or, at least, a model inconsistent with earlier or later models constructed on information from a different viewing situation or context.

However, there is a whole subset of viewpoints that can give rise to such illusory situations, with models inconsistent with the majority of other viewpoints. It is not even a matter of these 'illusory' viewpoints being less common. Rather there are two options:

  1. The illusory viewpoint may provide insufficient information to our perceptual system, which is unable to conclusively build an accurate model. Thus the most likely model is chosen, in the view of the brain's experiences. Many possible mental models are consistent with a single viewpoint.
  2. The brain's system for choosing a model may have insufficient experience to judge what features of the input determine the model. We may be young and innocent.
  3. The brain's system for modelling things is intrinsically flawed, and no matter how much experience of the rare, 'illusory' scenarios is given, there will be persistent inconsistencies in the models chosen by the brain.

If the brain had several possible interpretations of a single set of inputs, there are two possibilities. Either we know consciously that there may be many possibilities (say, [example]) or else we are unaware, and the brain autonomously chooses one model perhaps statistically (as in some optical illusions).

Against popular psychology

Popular psychology should now be set aside from true psychology. The question of how we think and make decisions is approached in different ways: true psychology aims to represent how we think and make decisions; whereas, popular psychology aims to represents how people structure, theorise and express, in real life, how we think and make decisions. Psychology is the study of thought, but popular psychology is the study of meta-thought. Making this distinction clear even at a theoretical level, will have impacts at a practical level upon our experience of reality. This change in perception is clearest in our interaction with other people, where, by distinguishing people’s actual motives from their post-hoc rationalisation of their motives, we can alter our own actions to have the best effect on their future behaviour.

For example, a man who completes a marathon because the alternative is to face a booing and overly sympathetic audience at the expense of the pretence of exhaustion, may believe he completed the marathon because of the strength of his ‘willpower’. The distinction between these ways of expressing his motives, while academic, carries subtle implications.

Willpower is an illusion

Willpower is a construction that may well delude the nature of the mind from itself. It may be that it is a functional construction, in that, it helps us in rationalising the apparently irrational, which is in turn is ultimately deterministic but on a level complex enough to defy analysis (at present), and ultimately draws on information unavailable to consciousness. In this view, willpower is a conceptual system inherited (partially socially, partially biologically) from our ancestors because of its advantage in generalising from idiosyncratic but successful decisions.

“Our bodies are meant to X” or “Our bodies are not meant to X”

Clearly, though it may be a useful recourse in dietetics, fitness or lifestyle, this statement leads to ultimate confusion. For example, our bodies were not meant for eating three large meat meals per day. 1) The statement cannot be true with sensible definition of the words ‘meant for’. Even given its truth, one cannot come to the implied conclusions that 2) three large meat meals per day is detrimental to the body or harmful to health, and 3) we ought or ought not to X.

1) We talk here of objects having a purpose, that is, the physical body being meant for a certain role. The given phrasing presupposes a world view in which there is reason behind bodily structure. That is to say, there are specific restricted goals that the body has been given the potential to fulfil; i.e., that there is design. It is not feasible to interpret the phrasing in a teleological way, unless we concede to a system that is homomorphic with creationism. But rather we should interpret it in a pragmatic way. But this pragmatic interpretation also is barren.

2) in evolutionary terms, our ancient ancestors were once single celled, and even with them we share certain sequences of DNA. As we progress through the generations, many different habitats and niches were explored; to all of these we were in the process of adapting to. The rate of mutation (and thus of evolution) has been tailored, coarsely, to match the rate of change of the environment we live in – that is, rather slow. And as a consequence, no organism is ever “fully adapted” to its current environment. Multiple vestigial components will always remain in the body, some of them perhaps remaining functional in the occasional but insignificant moment, but most of them essentially wasteful and dispensable. There is no saying whether, at a later date, such vestigial components might regain their use, and to some extent the latent ‘genetic code’ to produce such features could lie dormant. But at root, the dynamic, incomplete and inefficient nature of evolution prohibits us from even suggesting that we are best adapted for any single ecological niche.

3) needs no introduction and simply takes the form of a skeptical ‘is/ought’ argument.

One conclusion of all this is that we cannot look to evolution for guidance on lifestyle.

Religion as an example of ‘functional social fallacy’

Religion and popular psychology could both be categorised as social constructions that mould the mind from a young age, as evidenced by their cultural specificity and diversity. At the same time, they are both not direct logical consequences of the observed evidence – from our limited points of view, they could both fit with the evidence, but to infer them directly would be a logical fallacy. And indeed, they both appear to serve functions. The functional nature is not always easy to demonstrate, particularly in our current, non-anthropic, evolutionary mode of reasoning, in which the absence of a function indicates a flaw in our argument. Religion’s function has been discussed at length by other authors. Popular psychology’s function can be discussed if we imagine a world where it were absent.

Cognitive thought itself is an illusion


The reason for a model is prediction, useful prediction, which allows life-promoting improvements.

  1. The perceptual mechanisms can hone in on the features that are going to discriminate between possible future actions.
  2. The mechanisms of action can prepare to speed up response.

Many might say that reasoning is a faculty of all human minds. The truth is that reasoning relies on no abstract truth, but on the principle of allowing the statistically most efficient reactions to perceptions. One consequence of this interpretation is that speed outweighs accuracy in many thought-based situations. [Example] However there is one further limitation on the brain: it appears that only one consistent model can be computed at a time, so there can be no room for concurrent processing of two 'possible' futures.

I believe this to be a resource limitation. The enforcement results in what we count as illusions - the requirement to reduce the possibilities to a single 'reality': the reality in which we subjectively live.

Can we perceive two models of the world at once? We might say that conflicting models are mutually exclusive. What we really mean by this is, that 'conflict' has nothing to do with the information in a model itself, but on whether it predicts different futures. And the processing of the models is not fixed but acquired; so the idea of two representations being in mental conflict is experiential.


If the very idea of a conflict depends on experience, and may vary with time, can objectivity ever be attained? This depends on whether there is a common convergence point to which all mental processing systems tend as more experience accumulates.

The events we experience occur in very different orders and with different features in each individual on the planet. However there are many common elements in experiences which might determine common elements of model-processing and prediction, between individuals. Perhaps physics is an example of this.

What distinguishes duty?

Duty is not just any old kind of motive for action. It specifically implies an assigned, predetermined reason. It can be thought of as a 'must-do', 'must-follow' rule that can be used to govern the actions we perform. Duty must or should not be broken; that is how it is defined. Can this idea have any meaning at all in the framework of thought?

Duties are broken. From time to time there are lapses of the will that are said to constitute violations of the rules. In fact, without the possibility of a rule being broken, a rule has no effect whatsoever. In many ways, a rule never has an effect. It is merely a tool in describing the way things are, in a way that allows easy extrapolation to the future. It is based on experiences, and describes consistencies; specifically those consistencies which we could imagine not being there. Only because we could imagine a world where things did not fall to the ground, do we think of gravity as a constraint - as a rule that binds all matter - rather than an observation.

What effects to rules have? Indeed they do help us make some kinds of decisions; by using them to extrapolate from the present, we can predict the future and act in a certain way. Cognitive reasoning allows the weighing up of factors that grants an appropriate action.

Duty's social volatility

Against Rights

Communal will

There shall be a group of people, a committee, whose job is to compile a communal will. The communal will shall be invoked regularly and frequently, upon an issue that is decided by the committee. The frequency of invocation will be in the range of days to months. Each time, a question is posed and the opinion of the nation is sought. The opinion will be an important source of information for the government, and will provide it with direction.

The compilation of a communal will must involve the following:

  1. Selection of the subject
    The subject must be one that has been discussed previously by the government.
    It may concert an action that has already, or not yet, been taken by the government.
  2. Formulation of a question
    The question will be in the form of a dichotomy, with two possible choices.
    The question must be posed in an unbiased way.
    Example: Would you rather that the government did: A) ... or B) ...?
  3. Educating and informing the nation of the significance of the question
    This is perhaps the most important component of the will, and the most difficult.
    Each person over the age of 5 shall understand the question to a level appropriate to him or her.
  4. Local and national poll
    The question shall be followed by a scale between A and B. This may be numbered.
    Each person marks the scale to indicate how strong a preference the individual has.

The will is expensive to maintain. One result should be that each individual is involved in the government of the country. The will shall be funded by the government; once instated and working, the system should be tested by bringing to the people the question 'Is the communal will to be funded by the gover

In summary, nobody has capacity to make a decision except an ‘expert’ on that particular type of decision – where we generalise types of decision into categories depending on the kinds of information that are relevant to the decision (other than individual personal factors).

After all this, there are only a few ethical questions that remain unanswered:

Retributional vs rehabilitational justice; capital punishment


Value of potential life/possibility (temporal discounting in the future)

and scientifically, when is an individual aware?