Morality

Cognitive dissonance

“Extraordinary claims require extraordinary evidence”

– Carl Sagan (1934 – 1996)

These are excerpts of a discussion I had with one Mr. Kroor one fine evening when we had nothing better to do. We were debating over the talk Sam Harris gave at TED. This is my take on the talk.

The central theses of Sam Harris’ talk were:
(i) There are right and wrong answers to moral questions, and
(ii) Science can objectively tell us what is right or wrong.

Claim (i) betrays a stance of moral absolutism/naturalism. When one makes such sweeping statements, one should substantiate it with adequate evidence. Else they just remain claims rather than facts. Also care must be taken so as to not fall prey to Hume’s Guillotine (the is/ought problem, or I/O for short). Consider the following statements.

(i) Punching people in the face causes them pain.
(ii) One ought not cause people pain.
(iii) One ought not punch people in the face.

When you take a stance on a moral issue (like I do not want to cause people pain), you are being normative (read subjective). You have taken your stance based on what you as an individual think is right or wrong. In other words you are biased. Once you have taken that stance you look for facts; facts that tell you what causes pain to people (punching people in the face for instance). Notice that this is a factual statement and not a normative one. And this is where science comes in handy.

Finally, armed with scientific facts and the power of reasoning, one goes about their life doing what they feel is the right thing to do for any situation (stop punching people in the face because causing people pain is wrong or you may get punched back which will cause you a lot of pain). So the role of science is only as an arbiter in

(i) deciding the facts that carry the required import, and
(ii) checking for logical inconsistencies.

Harris says in his talk:

Values are a certain kind of fact. They are facts about the well-being of conscious creatures.

In the example above we saw that the factual statement (i) was distinct from the statement on moral values (ii). One cannot directly go from statement (i) to statement (iii) without invoking some form of statement (ii) and one can keep on adding similar normative statements to plug any leaks. And these statements are reflective of our biases and does not follow from (i). In other words, neither statement (ii) nor (iii) is a logical conclusion that one can derive from a factual statement. So a value statement like (ii) or (iii) are not facts in themselves.

In conclusion, facts on values or value judgements (like statement (i)) are not values themselves. And hence, science cannot give us right or wrong answers to moral questions since it deals with facts and factual claims. Science can give us facts about wellbeing of creatures. Right or wrong answers to moral questions differ from person to person (from claim to claim). These value statements which carry within them subjective biases hence cannot logically follow from factual statements (i.e., they cannot themselves be factual statements).

Kroor’s reply

Consider the following:

(i) I don’t like pain [Universally observed in animals. Humans, say, presently]
(ii) If I cause someone pain, they will retaliate [Almost universal in animals; hence in humans]
(iii) Therefore, I should not pain someone else. [Evidently normative prescription; arrived at from a universal observation and an almost universal observation generalized slightly].

Is this not what Harris is saying when he says “Values are a certain kind of fact. They are facts about the wellbeing of conscious creatures”? If yes, we’ve apparently proved him right. If not, you take objection to the golden rule or resort to the possibility that I’m a masochist. I think that (i) and (ii) hold quite generally.

Can moral statements have absolute answers?

I don’t know.

Can most of conventionally accepted morality be derived from (i) and (ii) above?

I think so.

So, can reason/logic give answers to moral questions, given (i) and (ii) above?

Evidently.

Have I shown that all morality is derivable inductively from the golden rule?

No. Abortion, for example. Are you hurting the foetus when you abort it? We don’t know. Will the foetus hurt you back, even if you are? Definitely not. Is abortion, therefore, murder? I don’t know.

A rebuttal

(i) I don’t like pain [Universally observed in animals. Humans, say, presently]

(ii)If I cause someone pain, they will retaliate [Almost universal in animals; hence in humans]

(iii)Therefore, I should not pain someone else. [Evidently normative prescription; arrived at from a universal observation and an almost universal observation generalized slightly].

Claims (i) and (ii) are normative and hence are ought claims. To say that animals (in this case you (we’re the fifth ape)) don’t like pain is an opinion (subjective) whereas a statement like pain causes suffering in animals is objective and factual. To say animals will/should retaliate when subjected to pain is again your opinion. Facts may tell you that very often they do. It does not mean that they always have to do which is a hasty generalization on your part. In short you have not given me any factual claims to consider your argument seriously.

Let me rephrase your argument for clarity.

(F1) Pain causes suffering in animals (fact).
(C)  I should not pain animals (conclusion).

As I said earlier, to reach the conclusion given above you need to make a subjective claim like (S1) I don’t like paining animals. This can indeed be reinforced by another subjective claim (animals are cute or torture is bad or I am a Jain etc). If you then now add another observed fact that (F2) animals often retaliate when tortured you need another subjective claim (S2) animals will torture me when I pain them. The two subjective statements can then be combined to read (S) I don’t like paining animals because they will retaliate if I pain them. There is simply no way to work around this problem. The subjective claim simply does not follow from facts.


Is this not what Harris is saying when he says “Values are a certain kind of fact. They are facts about the well-being of conscious creatures”? If yes, we’ve apparently proved him right. If not, you take objection to the golden rule or resort to the possibility that I’m a masochist. I think that (i) and (ii) hold quite generally.

I hope this is not what he meant. If yes then he has made the same mistake as you in not understanding the precepts of moral relativism and the I/O distinction. There is a reason why I/O problem has stood the test of time. But I think he’s too smart to make such a fundamental error. Waiting for further clarifications from his book.

Can moral statements have absolute answers?
I don’t know.

No as per current understanding.

Can most of conventionally accepted morality be derived from (i) and (ii)above?
I think so.

Circular reasoning. From the statements “Bible is the word of God” (normative) and the “Bible says that God exists” (factual) you cannot reach a conclusion that “God exists”. This is a classic example of begging the question. In other words the normative statement does not follow from the factual claim. I don’t know how many times I’ve had to repeat this.

So, can reason/logic give answers to moral questions, given (i) and (ii)  above?
Evidently.

Evidently no (see above).

Have I shown that all morality is derivable inductively from the golden rule?
No. Abortion, for example. Are you hurting the foetus when you abort it? We don’t know. Will  the foetus hurt you back, even if you are? Definitely not. Is abortion, therefore,  murder? I don’t know.

Golden rule is not a fact. It’s a choice (which you either choose to follow or not). Other than that I find no flaws there. Nice reductio ad absurdum to your own argument there! (see question above)



Epilogue

While I personally don’t condone immoral deeds, I definitely do object to the statement that science can be used to “derive” answers to moral questions. There are a lot of grey areas and the potential for abuse if such systems get implemented needs no mention.
Advertisements