On Imperfect Knowledge

This is a column published in this space two years ago, reprinted at the request of a reader who wanted to see it again.


Sean M. Brooks writes and asks:

In debates or discussions with other online friends, I’ve been told that “opinions” cannot be wrong, false, mistaken, erroneous, etc. My reaction was to argue this did not make sense. It could be my opinion that 2 + 2 = 5 or that Hitler was a noble, wise, saintly, and holy man. Are these “opinions” truly not wrong or false?

One person did concede an opinion can be factually wrong while still arguing opinions cannot be wrong. This did not make sense to me–and I rejected it as self contradictory.

If this interests you, do you have any comments to make? Am I wrong to say opinions can be erroneous or false? Am I missing something?

The short answer is that you are right and they are wrong, because if no opinions can be false or mistaken, then my opinion that some or all opinions can be false and mistaken cannot be false nor mistaken.

The long answer is more subtle: it depends on the meaning of the word “opinion”.

The longest and best answer requires a few paragraphs on the nature of human knowledge, and requires we draw some distinctions.

For the long answer, when in doubt, go to the dictionary.

(An older dictionary is preferable to a newer, since the rats of political correctness have infested many newer dictionaries, and simply and insolently give false definitions).

In this case the word “opinion” means:

1. a belief or judgment that rests on grounds insufficient to produce complete certainty.
2. a personal view, attitude, or appraisal.
3. the formal expression of a professional judgment: to ask for a second medical opinion.

So, under the first definition (which seems the way you used the term) a judgment or belief that rests on some grounds, even if those grounds produce limited rather than perfect certainty, can indeed be true or false, albeit reasonable men can differ.

That the sun will rise tomorrow is an opinion, albeit everyone but a philosopher will agree the statement “the sun will not rise tomorrow” is false, even if the grounds on which the statement rests do not convey the perfect certainty of conclusions of mathematics or logic.

And under the second definition (which seems to be the way they used the term) a personal attitude cannot be right or wrong, because it is merely the report of a subjective psychological state: My preferring pie to cake is an opinion, a matter of mere taste. I can neither argue I am right nor can you persuade me I am wrong on this point. However, the moment I can give a reason for that preference, as a man who says he prefers carrots to cake because carrots are healthier, the reason is open to debate.

The third definition is like the first: a professional opinion can be right or wrong, because a doctor can misdiagnose a case. That is precisely why we seek a second opinion. You interlocutors cannot be using the word in this third definition, since it would make nonsense of their argument: the reason and the only reason why we seek a second opinion in the case of a doubtful doctor’s judgment is because the imperfect knowledge of the doctor is (1) is imperfect and (2) is knowledge.

For the same reason, the only reason why we debate matters or reason about them is because our imperfect knowledge of debatable matters is (1) is imperfect and (2) is knowledge. On any point where our knowledge is perfect, there should be no debate; and on any point that is mere arbitrary preference, there can be no debate. The mere fact that a man engages in debate means that he regards the topic to be one of imperfect knowledge.

The opinion that opinions can be factually wrong but not be “really” wrong is a case of ambiguity. If your interlocutor was not merely spouting nonsense, one option is that he means the word “opinion” in the first sense (a matter of judgment) in the first half of the sentence and means the word “opinion” in the second sense (a matter of taste) in the second half. The other option is that he is engaged in Doublethink, and holds that an opinion can be factually correct but politically incorrect.

I was not privy to your conversation, but my hunch is that your interlocutors were confusing (1) whether the statement admits of being true or false (2) whether the statement is known to a current human being to be true or false.

For example, the statements “King Richard murdered the children in the tower” or “The ghost of Pompey appeared to Caesar” or “A dark star in the Andromeda galaxy has fourteen planets orbiting it larger in mass than Jupiter” or “There is a blue giant sun shaped like a cube in a galaxy in the Corona Borealis Supercluster” are clearly in the first category.

They are either true or false. Either Richard murdered the children, as he is alleged to have done, or he did not. Either Caesar saw Pompey’s ghost, as he is alleged to have done, or did not. Either there is indeed a star that emits no light in the Andromeda galaxy with fourteen satellites move massive than Jupiter or not. Either there is a cubicular star in Corona Borealis with the emission properties of a blue giant or not.

I myself do not know the truth of any of these statements. I may have an opinion based on my general knowledge of history, or of how ghosts act, or of how planets and stars are formed, but this general knowledge is less than perfect: it is theory, speculation, or guess.

If my imperfect knowledge is based on an assessment of the trustworthiness of a source or of a witness, then there are two levels of opinion at work: knowing the truth versus knowing the trueness. The first is my general knowledge of the subject matter; and the second is my assessment that the source on whose testimony I rely for my knowledge is a true and trustworthy source. The first asks whether the testimony is compatible with what else I know to be true; the second is whether the witness has earned my trust in other things, or provides some other indicia of trustworthiness.

For example, a man who for philosophical or theological reasons steadfastly refuses to believe in ghosts will reject the statement about Caesar and Pompey as surely false, even if the historian who reported the event is otherwise reliable and rigorous and honest. In that case the man’s opinion is weighing two different things of two different types: his philosophical beliefs about the universe and his assessment of an historian’s honesty.

In a case like that of the dark star in Andromeda, no one on Earth could even pretend to have eyewitness knowledge of such a thing, because the dark star would not be visible whether it existed or not, nor could any planets, with our current instruments, or any instruments likely to exist in the foreseeable future, be detected at such unimaginable and appalling magnitudes of distance. The statement is certainly true or false, but you or I have no way to guess which is which.

In the case of a cube-shaped star, we are on firmer ground. I can argue that there is no cube-shaped star in the Corona Borealis supercluster on the grounds that cube shaped stars would be pulled into a globular shape in the normal course of events. This is inductive knowledge, like my knowing the sun will rise tomorrow. It is conditional knowledge, like the vast majority of all the knowledge in our heads, based on our model of the universe.

One can almost express this in terms of an if-then statement: IF my knowledge of the cosmos is tolerably accurate THEN no cube shaped stars can exist anywhere.

Or one can express this in terms of a cost-benefit ratio: the cost of my entertaining the belief that a cube shaped star could or does exist is that my faith in my model of how the universe works must be dismissed as not just wrong but grossly wrong. The more outrageous the statement we are asked to swallow, the more of our previous worldview we are asked to dismiss, hence the higher is the cost of belief.

Compare the “cost” of believing in a dark star with fourteen planets (something that could be true in our current understanding of astronomy) with the “cost” of believing in the Bizarro star (something that, if true, shatters more of our understanding that Einstein shattered when he updated Newton).

Let me pause to mention that a more elaborate hypothetical, such as one which says the cube shaped star is suspended equidistantly between eight powerful gravity sources that pull it out of shape, or says that it is an artifact constructed by intelligent aliens, would have a lower “cost” since it would do less damage to our current understanding of the universe.

The human mind is habituated automatically to seek such cost-lowering hypotheticals, which is why people so quickly grasp at ad hoc explanations. It demands less cost, for example, for a communist to believe that the interference of evil international foreign powers prevented Soviet Russia from awing the world with the efficiency of her wealth-production rather than to revise the whole Red theory and admit that communal ownership is inefficient, much less than to admit that Communism is not an economic theory at all, but an Millenarian voodoo-deathcult belief.

Let emphasize that the habit exists because, outside of the temptation to ad hoc errors, it is generally beneficial. When plate tectonics were discovered, the belief that all continents once were one super-continent, until then too hard to believe, became believable. Something that seemed impossible now seemed possible, even likely, because a mechanism had been discovered to explain how such a far-fetched thing could be.

Knowing that the Bizarro Star does not exist is not perfect knowledge, like the knowledge that twice two is four.

Perfect knowledge consists of those things which must be true or else the laws of logic are wrong. Imperfect knowledge consists of those things which must be true or else something basic about our understanding of the universe is wrong, such as that the law of gravity has a hitherto unsuspected exception. It could in theory be wrong: one can speculate in the hypothetical about a cube shaped star.

One cannot speculate in the hypothetical about an alternate number line where twice two is not four, because the laws of logic which make twice two four are the same laws which we use to speculate with. It is unimaginable. By that I mean: It is not just unimagined. A thousand-sided regular figure called a chiliagon is unimagined. I have never heard of a human who can picture such a thing in his head. This is something else. Not even a Martian or an Elf or an Angel can imagine twice two not being four.

In other words, any speculation that involves a repudiation of the basic rules of logic, the basic rules of reason, cannot be reasoned about. The “cost” is for all practical purposes infinite.

All the above takes the question of your interlocutors to you seriously. But whether the question should be taken seriously depends on their intent, which I cannot, from my imperfect knowledge, guess.

You see, your interlocutors sound unfortunately very much like Lefty nihilists who seek to defend the indefensible by the sloppy tactic or trick of declaring certain topics to be beyond debate.

The sloppiest of the many sloppy Lefty tactics or tricks for silencing honest debate is to decree their pet topics matters of mere opinion, not subject to the rigors of scientific or logical thinking, and therefore inviolate. One can easily spot that it is a trick and not an honest opinion by bringing up a topic where they feel strongly, let us say, the “opinion” that women should neither vote nor own property, or the “opinion” that all gays should be stoned to death, and all uppity Negros hanged by lynch mobs in white hoods, all Jews sent to the ovens, and all poor people enslaved and starved, and then see whether their firmness in declaring the matter to be one where there is no right nor wrong remains firm.

These and many other examples show that there are indeed certain things we know and that we cannot pretend we do not know. The very pretense, the complex excuses, or justifications, or wrath, or pride, which we must use to whitewash and mask our dirty sins shows that we know them to be sins and that we cannot escape the knowledge.

No one thinks and no one can think that the Nazi holocaust of the Jews was a morally neutral act, one where it is merely a matter of opinion whether or not the act was right or wrong. “I prefer pie to cake, you prefer Jews be slaughtered cruelly like beasts in unimaginable numbers: one preference is the same as another, merely a subjective opinion.” No one says this sentence and means it seriously.

Those who say such things cannot mean them seriously, because if they were serious they would be agitating for the abolition of all human law. Human law rests on the idea of moral responsibility. We do not put babies or dogs on trial, nor when a tree falls and kills a lumberjack do we haul the tree to a hanging tree and hang it. For that matter, we do not put madmen on trial for the single reason that the defect of their reason makes them incapable of moral responsibility, no more than a baby or a beast. The argument that all moral judgments are matters of opinion as vapid as preferring pie to cake, a mere matter of arbitrary and frivolous preference, is even worse than an argument stating that all men are madmen unable to carry out moral responsibilities: for it is an argument that moral responsibilities do not exist at all, because morality does not exist at all.

Obviously no one believes this, for if he did, he could not honestly disagree with any argument contradicting him, because honesty presupposes morality.

For let us suppose, for the sake of hypothesis, that two men were in the universe where, as a law of nature, it is the case that all statements are mere opinion and hence there is no truth and no moral imperative to speak truthfully.

If the first man says, “All things are mere opinion: hence, there is no such thing as truth and no moral imperative requiring us to speak it” and a second man who agrees in his heart with him stands up and, knowingly and deliberately lying with malice aforethought, says, “I disagree! We must speak the truth!” what shall the first man say back? “That is a malicious untruth! How dare you violate the moral imperative!”

Ergo if we lived in a universe as described in the hypothetical, then all debate, hence all reasoning, is vain, and we could not even determine this much: whether or not we indeed live in such a universe as in the hypothetical.