The Will-Ought Distinction

David Hume, the originator of the is-ought distinction

Posted by Thomas Scarborough

The is-ought distinction is a philosophical classic which lies at the core of moral philosophy. It states that one cannot derive ‘ought’ sentences from ‘is’ sentences -- in other words, one cannot derive ethics from fact. This rests in turn on the things-relations distinction. While ‘is’ describes how things are related to one another, it is unable to describe how they ought to be related.

Instead, here, I seek to introduce another distinction which, as far as I can see, is non-existent in the world of philosophy  -- yet may be more important than the is-ought distinction.  The goal here is to present a will-ought distinction. It is offered in broad outline, while acknowledging the fact that there are many nuances at play (that is, what follows is necessarily simplistic).

Now ‘is’ and ‘will’ play a large part in science.  The scientist begins with ‘is’, which is the facts -- or the pieces of what is out there -- and these facts, for a scientist, may not make much sense in the beginning.  As scientists seek to understand the facts, then, they play with them -- they rearrange them, reword them, have a few drinks together as they argue over them -- even bet on hypotheses, and so on.  In other words, the ‘is’ of the matter may be a fairly detached activity.

This is the first of the (if one so prefers) four parts of the scientific method: characterisations, hypotheses, predictions, and experiments -- and the first of these four, namely characterisations, is about observations, definitions, and measurements of the subject of inquiry.  In other words, it is about establishing the facts.

Yet facts alone offer little or no explanation of phenomena.  There comes a point where these facts must be combined in such a way as to present a theory.  At this point, ‘is’ (the facts) becomes ‘will’ (the theory).  Theory is, after all, that which will happen.  Theories predict things.  If they do not predict things, they are not theories.  We now find, besides ‘is’ and ‘ought’, a third category.  In the context of science:

• ‘is’ refers to fact,
• ‘will’ refers to theory,
• while ‘ought’ refers to value

Let us now notice that ‘will’ and ‘ought’ both have to do with expectation.  Both may be defined -- at least in a great many cases -- as ‘reasoned expectation’ -- and reasoned expectation is a fairly standard definition for theory.  One cannot call theory reasoned certainty of course, as every scientist will freely point out.  Neither is this true of ‘ought’.

Further, both ‘will’ and ‘ought’ may refer both to personal and impersonal things.  This fits well with the things-relations distinction above.  They both have to do with how we expect things to be related to one another.  Things ‘will’ be so and so ordered, or they ‘ought’ to be, whether this refers to a person's behaviour (‘She will ...’ ‘He ought ...’) or to the arrangement of objects, events, or concepts (‘It ought ...’ ‘That will ...’).

How, then, should we distinguish ‘will’ from ‘ought’?

Unlike ‘will’, ‘ought’ may often be non-normative -- which means that one may not seriously, in every case, expect it to happen.  As an extreme example, someone might say, ‘There ought to be two moons orbiting the earth’ (which is impossible).  But then, too, someone might consider, ‘Crystal Palace ought to win the Premier League’ (possible, though not likely), ‘The rocket ought to land safely’ (likely), or 'This experiment ought to produce an alkali’ (all but sure).  There is a continuum of additional information, therefore.

Now I need to make a further distinction.  In that case of proposing two moons, we might really have meant, ‘There ought to have been (perfect tense) two moons.’  A great many ‘ought’ sentences may be interpreted in this way.  Subtract such examples from all examples, and this increases the number of oughts which are more seriously about the future.  ‘Ought’ now comes a lot closer to ‘will’.

Perhaps one might argue, too, that ‘ought’ may be distinguished from ‘will’ by seeing the term as applying only to ethics -- and not to such neutral subjects as those of science.  Yet how should one make such a distinction?  It is artificial -- not to speak of old-fashioned.  I have said that ‘ought’ represents the way that things ought to be related to one another, and this does not in principle separate ethics from science, or from any other activity for that matter.  This brings two important questions:

•  Is science in reality a language of ‘ought’? (Which is, all about value?)
•  Can one really separate ‘ought’ from ‘will’? (That is, to separate moral acts from any acts at all).

And one last question too:

•  Is every move we make a moral one?

Fact and Value: The Way Ahead

Grateful acknowledgement to Bannor Toys for the image

Posted by Thomas Scarborough

Philosophy may begin to solve a problem as soon as it has identified it.  All too often, it has not.  This post, then, is about defining a problem—no more.  It is one of the most urgent problems of philosophy.

One of the most important aspects of philosophy is ethics.  Yet there is an issue which is prior to ethics, which has to be addressed first.  It is the problem of the fact-value distinction—a problem which, since it first appeared on the philosophical map, has cut a divide between fact and value, and more importantly, philosophy and ethics.  In the words of Ludwig Wittgenstein, ethics has become ‘what we cannot speak about’.  Yet ethics is all that we do, from morning until night, from year to year.  Today, this problem has filtered through to the common person, and has caused profound disorientation in our time.  On a social level, we are conflicted and confused with multiple ethics, while on a global level, our ethics increasingly seem to have come apart, with widespread poverty, social disintegration, and environmental destruction. 

It seems easy to describe the philosophical problem, yet far from easy to offer a solution.  Should I take a walk in the woods today, or should I write letters instead?  Should I be a ‘bachelor girl’, or should I marry Joe?  Should we travel to Mars?  Should we drop the Bomb?  On the surface of it, our reasons for choosing one course of action over another might seem obvious, yet it is not something we find ourselves able to decide on the basis of facts.  The problem is basically this: we know that this is how the world ‘is’—yet how should we know how it ‘ought’ to be?  The philosopher David Hume gave the problem its classical formulation: it is impossible to derive an ‘ought’ from an ‘is’.  It is impossible to establish any value amidst an ocean of facts—and on the surface of it, Hume would seem to be unimpeachably right.  The facts cannot tell us what to do. 

As we seek a solution to the problem—because we must solve this problem if we are to find our way through to any discussion of ethics—Hume’s conclusion would seem to mean only one of two things: either he identified a problem which cannot be solved, or he was thinking in such a way that he created his own problem.  What, therefore, if Hume laid the very foundation on which the fact-value distinction rests? 

Hume considered that all knowledge may be subdivided into relations of ideas on the one hand, and matters of fact on the other.  That is, one begins with a handful of facts, then relates them to one another.  It is the simple matter of a world where facts exist, and these exist in a certain relation to one another—yet one finds no basis on which to determine what that relation ought to be.  Generations later, the philosopher Bertrand Russell wrote that many philosophers, following Kant, have maintained that relations are the work of the mind, while things in themselves have no relations.  While Russell was not saying precisely the same as Hume, he was not far off.  A similar view is reflected in the theory of language.  The philosopher Rudolf Carnap considered, in the words of philosophy professor Simon Blackburn (specifically about the ‘material mode of speech’), ‘Speech objects and their relations are the topic.’  Wittgenstein, too, held this view, in his own unique way, through his multiplicity of language-games.

A pebble is a thing.  A house is a thing.  Even gravity, ideology, taxonomy are ‘things’ in a way (we call them constructs), which in turn may be related to other things.  In a sense, even a unicorn is a thing, although we are unlikely ever to find one.  Things, then, may further be involved in what we call truth conditions—which means that they may be inserted into statements, which can be affirmed or denied.  And when we affirm such statements, we call them facts.  For example, we insert the thing ‘pebble’ into a statement: ‘A pebble sinks’—or we insert the thing ‘unicorn’ into a statement: ‘The Scots keep unicorns.’  Our things are now involved in truth conditions, which means that our world is filled with facts.  And if not facts, then denials of  facts. 

Here, I think, is where the problem lies—and the way ahead.  To say that there is a fact-value distinction means that we have first divided up our reality into things on the one hand, and relations on the other.  On what basis, then, might we find our way back to a ‘grounded’ ethics?  Personally I believe the solution lies in the direction of levelling both fact and value to value alone—or things and relations to relations alone—in all fields, including science and mathematics.  Yet even then, we would not finally have reached the goal.  Even if we should be able to see everything in terms of value, which values should then be true, and which false?  And having once solved which values are true, we would need to establish on what basis I should—or could—submit to them.

How Shall We Re-establish Ethics in Our Time?

Posted by Thomas Scarborough

We have nothing to show today, writes Simon Blackburn, for ethical foundations. At the beginning of the 21st century, we are ethically adrift. Not reason, not religion, not intuition now seem adequate for grounding our behaviour.

Not only does this present us with a philosophical problem. On a social level, we are conflicted and disorientated with multiple ethics, while on a global level, our ethics increasingly seem to have come apart – with deepening poverty, social disintegration, and environmental destruction being the order of the day.

Philosophically, it was David Hume who first observed that we cannot derive an 'ought' from an 'is'. We cannot, on the basis of a handful of facts, derive any values. Gradually, over the following centuries, this idea took hold – until, with Ludwig Wittgenstein, it achieved a wide acceptance. 'Whereof one cannot speak,' he wrote, 'thereof one must be silent.' He was referring, most importantly, to ethics.

How, then, shall we re-establish ethics in our time?

Many, today, would consider the very question to be foolish. Yet given two simple conceptual prerequisites, I propose that we may indeed re-establish ethics, philosophically:

• Prerequisite 1. The fact-value distinction would have to be set aside. This, I believe, should be possible by ridding ourselves of facts – and with facts, of things (facts are, after all, about things). This may not be as difficult as it seems.

• Prerequisite 2. We would need to reduce our world to relations, and only relations – without the existence of facts or things. Ethics – together with all other fields of inquiry – may then be defined in terms of relations, and relations alone.

With regard to the fact-value distinction then, it is important that we first understand that this rests on Hume's notion that all knowledge is subdivided into relations of ideas on the one hand, and matters of fact (and things) on the other. On closer examination, however, we find that this view cannot be sustained.

It was Francis Bacon who observed that the definitions of words have definitions. Similarly, we know that the features of words have features. Words, I have myself proposed, represent 'relations within relations'. Given that this is true, there can be no self-contained 'things'. Rather, our words (and thoughts) reach into an infinity of relations, and we speak about 'things' only by way of a truncated shorthand.

If, then, there are no self-contained 'things' in this world, and if our words (and thoughts) reach into infinity, then this must mean that we cannot speak truly about our world on the assumption that it is a closed system, in which we begin with axioms, origins, or specific reference points. This flies in the face of what we have before us. Rather, we need to step back from all systems, to view our world as an infinite canvas of relations.

I propose that we shall find, when we do this, that relations as a whole possess certain features – call them meta-features – many of which may only be recognised from a bird's eye view.

These features hold the promise of new ethical foundations:

• Feature 1. As we survey the infinite canvas of relations, we realise that some of the relations which we trace in this world do not in fact exist – and those relations which do not exist cannot form the basis of an acceptable ethics. Falsehood and deceit are, at bottom, attempts to trace non-existent relations.

• Feature 2. Since relations represent an infinite canvas – and any containment of relations detracts from this – the relations which we trace should range through all the world. The more we limit the scope of the relations which we trace, and the more we view things in isolation, the more we are at risk of fault or shipwreck, in any field.

• Feature 3. Since the relations which we trace should range through all the world, our ability to trace relations should not be obstructed or manipulated. This means that secrecy, propaganda, and misrepresentations are ruled out. Above all, an ethics of relations would favour an open society, since openness is a prerequisite for arranging our world.

• Feature 4. Relations are infinite, yet our minds are finite. Since our minds are capable only of encompassing limited regions of relations, this means that there will inevitably be relations which lie beyond our power to explain – and beyond our control. We need therefore to be acutely aware of our limitations of thought. There is no place for hubris.

• Feature 5. Infinite control is required to conquer infinite relations. Therefore our limitations may create within us powerful totalising urges. In view of our limitations therefore – which preclude any final ability to master our world – any ethics should avoid such totalising tendencies.

• Feature 6. When we elevate any given values above others in an infinity of relations, our (justified) fear that such values are empty and illusory grows. This may lead to a dysfunctional ethics, which drives us to fundamentalism – where fundamentalism is resistance to the fear of finding one's foundations to be exposed as baseless.

• Feature 7. We need to consider that we live in a world where new relations are continually emerging, which did not exist before, and in many cases were unimaginable just so many years ago. This means that ethics does not and cannot stand still.

Finally, when we view our world as an infinite canvas of relations, and only relations – having banished both 'facts' and 'things' – we may now define fact as those relations which are as we think they ought to be, and value as those relations which are as we think they ought not to be. In both cases, then – in all cases – we speak, at bottom, of 'ought'.

With careful consideration, we come to see that all of the natural and the human sciences, and all of our common life, represent value, not fact. Thus we escape the albatross of the fact-value distinction, and bring back ethics into the fold of philosophy.
____________________________

For a deeper exploration of these themes, one may refer to the author's Metaphysical Notes.