Are there such things as scientific facts? Does it make any sense for there to be? The answer is yes, but most of the claims which are assumed to be facts are not actually facts at all, at least not if you are speaking within the realm of scientific investigation.
First, let us consider a few definitions of the word “fact.”
- A thing that is indisputably the case –
- “A thing that is known to be true, especially when it can be proved” –
- “A truth known by actual experience or observation” –
So we have facts as being indisputable, known to be true, and things which are known by actual experience or observation. But in science, many facts are said to be “provisional truths.” This seems odd. So let us take a look at how science even works.
Ever since Karl Popper, science has been a process in which deductive reasoning is used to falsify models, using empirical evidence. All theories are just models which can be used to determine the nature of an outcome for a given experiment. We can say that an outcome occurs with probability 0 (is impossible), occurs with probability between (0, 1), i.e. is possible, or occurs with probability 1 (is necessary).
We have a large collection of models (theories) that we use. So what can we say is a fact in science? We can discuss facts in relation to a collection of models M. We can say that a claim is either consistent with M or not consistent with M. Why is that a fact?
We have concluded that a fact is something which is indisputable. In other words, regardless of any additional information, so long as people no how to evaluate the truth value of the statement, the result is the same. That is quite possible in the system above. Any scientific can evaluate a claim and determine whether or not the claim is consistent with M or not, and regardless of any additional information, the answer will be the same, so long as we are using M, rather than some other collection of models.
Observations and Inferences
We can also include direct observations, since facts can be considered truths known through experience or observation. The problem is, most of what we think of as observations are really inferences based on observations. Suppose you plant a sapling in a park. Six months later you go to the same park and see a tree, in the same spot that you planted your sapling, that is four times as tall. Did you observe that your sapling had grown? No. You saw a before and after state and used inductive inference to come to your conclusion that the tree had grown. So is it a fact that your tree had grown? No. It is, however, a fact that your inductive inference is consistent with our current models in botany.
Now, when scientists speak to one another, do they need to address model consistency when speaking of facts? No. It can be implied. If I say that “x is a scientific fact” I mean that it is consistent with our current theories. But people must be aware of the distinction. Because if I just say that “x is a fact” it becomes something which, at least in the minds of some, becomes indisputable, in its own right.