Our recent discussions, particularly on the thread about Jonathan Haidt’s response to Sam Harris’s challenge, left me thinking about the various scopes of objective facts. In retrospect, it’s a bit obvious to me now that a key question in moral philosophy is, if morality is objective, at what scope is it objective?
Haidt used the terms anthropocentric and non-anthropocentric to divide facts between natural ones and ones within the scope of the human species. He then uses the mouthful “emergent culture-specific anthropocentric truths” to describe culture related facts.
Just for my own mental categorizations, I’m using slightly simpler terminology. For one thing, “anthropocentric” feels like a term that is likely to require lengthy explanations every time it is used. Instead, I’m thinking of the following hierarchy.
1. Universal facts
These are facts that are determinable through the natural sciences, such as E=mc2 or F=ma. Universal facts are, as far as we can determine, true throughout time and space. This is the domain of fields such as physics, chemistry, and geology.
2. Biological facts
These are facts about how evolution and biology works, at least on Earth. I debated whether this is really part of 1, but decided that since we haven’t observed life anywhere other than on Earth, it might be prudent to consider this as separate. We may well discover someday that extraterrestrial life operates radically different than Earth life.
Even limiting ourselves to Earth, I also fully realize that the scope of biological facts is very vast, and what is true for certain types of life is often not true for other types.
This is the domain of biology, paleontology, ecology, and related fields.
3. Human biological facts, or facts of human nature
This includes facts such as the color blue, that we get hungry if we don’t eat, that we feel fear and other emotions under certain circumstances, and other facts about humans that are not due to choices we make.
Human biological facts are only true within the scope of the human species. This area is studied by medical science, evolutionary psychology, paleoanthropology, and related fields.
4. Cultural facts
This includes things like the relative values of gold and silver, humor, or the way we feel about gay marriage. It is studied by psychology, sociology, economics, and other social sciences, as well as history.
Each of the domains above is emergent from the previous one. However, while human cultural facts may reduce to universal facts, it’s really not productive to think of them in that way, since in all practicality we can’t use any of the universal facts that we know to really predict or understand human cultural facts.
So, where in this does morality fall? Well, along with Haidt, I tend to think that specific moral rules exist in 4. However, these rules arise from instinctive impulses, from programming, from foundations in Haidt’s terminology, that exist in 3. So from a descriptive viewpoint, morality is relative, although it’s grounded in, and constrained by, biological evolutionary facts.
Of course, the problem is that even the relative strengths of the foundations themselves are relative to our individual biologies, to our particular genetic and phenotypic profile. It means that our deepest feelings about right and wrong may simply be different. Conscience itself is, to some extent, relative to each individual.
That said, there is enormous overlap among us. Enough that it’s possible to form societies, and enough that we can study human nature in general and discover what the vast majority of humans will instinctively prefer.
We can scientifically study what people or cultures prefer. We can study why they prefer what they prefer. We can scientifically study the effects of those preferences on specific rigorously defined indices. We can establish objective facts about these things.
What no one has found a way to scientifically study, what we can’t really establish objective facts about, is which preferences we should have, which values we should hold. For example, what study could we do that would ever tell us whether gold should be valued more than silver? Or whether it is right or wrong to kill a deformed infant? We can’t, at least not without reference to some other pre-existing value.
These are value judgments. And as I’ve said in multiple posts, we ultimately don’t hold values for any cognitive reason. We hold them as a result of evolved programming, of instincts, of the moral foundations in Haidt’s theory.
- Jonathan Haidt: Why Sam Harris is Unlikely to Change his Mind (3quarksdaily.com)
- 7 Reasons Why It’s Easier for Humans to Believe in God Than Evolution (motherjones.com)
- Moral values aren’t absolute, but aren’t arbitrary either (selfawarepatterns.com)
- Moral Tribes: Emotion, Reason, and the Gap Between Us and Them by Joshua Greene – review (theguardian.com)
21 thoughts on “The scope of objective facts and morality”
0. Logical necessity
Some facts could logically be no other way, such as the mathematical fact that there is no greatest prime number. These belong to a level even more fundamental than universal facts.
Good point. As you’ve explained to me before, logical and mathematical facts can exist that have no correspondent entity in the world.
Great post, it is certainly a valuable pursuit to lay out hierarchies of facts. But I am left to wonder: why pursue knowledge of any of these facts at all? I think it is because there are evaluative facts that underlie the structure of knowledge in that it gives reason to posit structures of knowledge. I think there are brute evaluative facts that, if not in the category of universal facts, belong in their own category at the same foundational level. Why should we care if our beliefs about the world are correct; why is the pursuit of scientific knowledge worthwhile? The only explanation can be axiological: it is better to hold true beliefs than false beliefs. This basic normative truth, I think, also explains why we think lying is, in most instances, morally wrong. But it seems like we have this evaluative belief for cognitive reasons; we can see in our intellectual intuitions that it is better to have true beliefs.
Can we scientifically study why man desires scientific knowledge of his world? Perhaps we can find scientific explanations rooted in evolutionary and social theory to explain why man wants to know whether certain tokens of his doxastic rationalizing our reliable, but it seems doubtful that empirical sciences can tell us why we care about the empirical sciences.
I don’t think axiological facts are facts at all, or if they are they are only so like moral values in a level 4 or level 3 way. We find it is better to hold true beliefs than false beliefs because we are hard-wired by evolution and education to hold this value, not because it makes any objective sense to say that one thing is better than another.
We’re hardwired by evolution and education to think that it is better to hold true beliefs rather than false beliefs? While such phenomena might instill in us first-order evaluative beliefs, it seems quite strange to assert that evolution creates abstract/second-order beliefs. Perhaps education does, but it seems there would still need to be a point of origin for such a belief if it is not an a priori/universal truth, and as I just noted, evolution doesn’t seem able to provide such an origin. If you think it can, I would enjoy you sharing the theory that supports such a claim.
I would say that much of what we are hardwired to do by evolution arises not because of direct selection but as inevitable knock-on effects of other things more directly selected for.
So, for example, evolution selects for organisms which are successful eaters. Evolution therefore selects for organisms which like food and which develop successful techniques for finding food. Having beliefs about how to get food forms part of the toolset of intelligent food-seekers such as ourselves. However, beliefs can be false. It stands to reason that evolution would select for organisms that seek to minimise the risk of holding false beliefs that would not help to find food. Valuing truth is one tool which achieves this.
Therefore, as a second-order effect of selecting for successful food-seekers, evolution has indirectly selected for valuing truth. This is reinforced by similar arguments for seeking mates, seeking shelter, etc.
But that’s just a just-so story, albeit a plausible one. My point is that we are what evolution has made us. Whatever we value is ultimately a result of this evolution, whatever the correct just-so story might be that accounts for it. Cultural effects such as education also come into play, of course, but these also ultimately arise out of biological evolution, especially where human values appear to be universal.
Excellent, well put, I’m embarrassed now that I didn’t anticipate that; it is quite plausible, as you note. Taking what you said together with SAP’s ideas, we might say that evolution in some ways results in the instinct for true beliefs, and later on in cognitive development we theorize that truth is abstractly valuable. Fair enough. My concern, which I’ve noted to SAP in the past, is that I think evolution-focused views provide good evidence that instincts are the basis for our motivation to act in accord with value judgments, but it doesn’t necessarily follow from such a thesis that instincts constitute/create values.
Thanks ausomeawestin. I like this discussion you and Disagreeable Me are having.
I think we can definitely hold values that come about from reasoning, and I suspect the desire for scientific empirical investigation is one of them. But that reasoning comes about due to more foundational values. Those values may themselves come from reason and even more foundational values, but eventually you hit a brute layer of innate desires.
Of course, figuring out which values are reasoned, and which are grounded in human instinct, is far from an easy task. It’s why evolutionary psychology is so difficult, and controversial.
Even when a desire is innate, it’s very easy to fool ourselves into thinking we have reasons for that desire, that value, when in reality what we have are rationalizations. It’s also easy to fall into the reverse trap, assuming some observed trait is innate when it’s actually cultural. The “girls prefer pink” controversy comes to mind.
Very interesting. Just taking a guess, what kind of innate desire could ground the evaluative belief that it is better to hold true beliefs than false beliefs? An innate desire to be correct? I can certainly admit that there is plausibility to the thesis that our normative (first-order) moral beliefs are created by evolutionarily instilled instinct, but it really does seem absurd to me to say that evolution caused in us second-order beliefs — that is, beliefs about our beliefs. I certainly recognize the difficulty it making such hypotheses, but I think the onus is on evolutionary psychologists to offer such theories if they are going to challenge the value base for our beliefs as ontologically superfluous.
I think the onus is always on someone who puts forth a theory to find evidence for it. But I don’t think it’s controversial that we have instincts (hunger, mating, etc). The controversy starts when we claim a particular impulse, that is not seen in non-human animals, is an instinct. But if we’re wrong about a particular impulse being an instinct, that would mean it came about from reasoning in pursuit of some other instinct.
If not, then what’s the missing component?
Cognitive recognition of abstract synthetic a priori evaluative propositions?
I wonder if you could expand on what you mean by “abstract synthetic a priori evaluative propositions”. I understand each adjective, but can’t compute what you mean by the whole phrase.
That’s a fair question, sometimes I throw that phrase around without thinking about whether it is truly applicable to the discussion at hand; but I think it’s important here. The idea is that some a priori truths are not analytic in the sense of being immediately obvious to persons who understand the terms involved (i.e. a bachelor is an unmarried male is the common example). Synthetic a priori truths are truths that don’t have this immediate obviousness, but which, after carefully reflecting on the terms involved, sometimes for centuries, are seen to be such that there is a necessary identity between one term and another in all possible worlds. Hilary Putnam and Saul Kripke have argued that because what we call ‘water’ is causally regulated by a certain molecular formulation, ‘water’ is causally regulated by H2o in all possible worlds, such that it is a synthetic necessary identity, and thus, a priori truth, that ‘water’ is H2o. I think it is conceivable that we uncover, through reflection, which natural properties (harm causing, beneficence showing, etc) are coextensive with moral terms due to synthetic necessary relations between natural non-moral properties and the moral non-natural properties that supervene on them. If we can uncover such necessary identities as being causally regulated in all possible worlds, it should lead us to conclude that moral values are not constituted in any way by our evolutionarily developed instincts, as evolution could have resulted in different instincts, and thus evolution entails that those identities are not synthetic a priori necessary relations.
So the missing component that you inquired about earlier in this thread, I think, is synthetic necessary identities between non-evaluative natural properties and evaluative non-natural properties. That evaluative properties supervene on non-evaluative properties shouldn’t lead us to give evaluative properties any less epistemic importance as foundational axioms of reason.
This is a really good discussion. I like SAP’s attempt to categorize different kinds of facts, although Disagreeable Me’s point that there are logical/mathematical facts suggests that facts don’t necessarily belong to a simple hierarchy. Although we might agree that logical/mathematical facts are facts in all possible worlds, whereas the ones SAP listed aren’t.
That raises the question whether, if something like AusomeAwestin’s “True beliefs are better, all things being equal, than false beliefs” is a normative fact, is that a fact in all possible worlds too? It would seem to be, if it’s a fact.
There do seem to be some foundational norms, as SAP says, but I’m not sure it’s appropriate to call them “facts”. And yet I’ve never been able to say exactly why they’re not facts. I usually end up simply thinking there’s something different between “what is” and “what should be”, or between the “real” and the “ideal”.
Why can there not be “what is about what should be”?
“What is about what should be” is perfectly ok, although a little verbose. The problem is that it’s just another way of saying “what should be”! Grammar aside, the point I was trying to make was that there seems to be a fact/value distinction, even though the distinction isn’t perfectly clear-cut and it’s hard to explain or express in other terms (for me anyway).
Lolz quite right, quite right. My point was really just to imply that evaluative propositions can be truth-apt. I’m not sure what to make of the fact/value dichotomy. I thought Putnam and others had effectively defeated it, hence the disintegration of logical positivism. So either there is no fact/value dichotomy, or the distinction is irrelevant to there being objective truths in morality as the truth-aptness of moral/evaluative propositions can secure the possibility of objectivity in ethics.
It seems to me that the existence of the norm is a fact, but not the norm itself. So, us believing that survival is good, is a fact, but survival being good, isn’t.
Of course, if you add in a “for”, it can become a putative fact. Whether or not spanking kids is good or bad is not a fact. But spanking kids can be good or bad for minimizing their tendencies toward violence in later life. Now we have an empirical question. Of course, whether or not having a tendency toward violence later in life is another value judgment, so we’ve just moved the brute value back a bit.
I like your way of putting things here. So if ‘fact’ is defined in such a way that it excludes values, but evaluative propositions can be truth-apt in the sense that such propositions can be true or false, can we not gather empirical knowledge of values? I don’t think we can, due to a “is/ought” gap. This suggests that the fact/value dichotomy rests on a distinction between a priori and empirical knowledge, and that knowledge of the fundamental evaluative beliefs can only be known through cognitive processes of rationality, not empirical study of instincts.
Not quite sure how the fact / value dichotomy rests on the distinction between a-priori and empirical knowledge. It seems like second order values can arise from both, but that first order, or primal values, simply exist without reference to either.
I agree with you that second-order values might arise from both a priori and empirical knowledge, and primal values from neither, I think, historically, the fact/value dichotomy was very much based in separating a priori pursuits from empirical pursuits.