Amateur here.

I always held that the only way to identify truth is by the scientific method. You test it. But, as I sit here smoking my pipe, I think there has got to be some method to deal with real world uncertainty. I'm not arguing against the scientific method. I'm asking a different question. I'm asking about how to deal with uncertainty in an uncertain world and what practices do you adopt to navigate towards the truth and away from suboptimal conditions. I am hoping that thousands of years of philosophy has the answer or at least a few contenders that a thoughtful human can use today.

Put another way, what do the great philosophers say about the optimal way to think? I have interests in machine learning so this question is important in that regard also.

Also, I'm aware of reinforcement learning and statistical models which work very well. But, you still have to put these things together the right way. I'm looking for some other insights that philosophy might be able to offer.

  • 1
  • Asking after "the optimal way to think" does not strike me as particularly constructive here.
    – Joseph Weissman
    Commented Aug 13, 2011 at 16:18
  • 1
    I've been trying wrap my head around this question for a while now. It seems you are asking how to quantify and deal with uncertainty. But this covers so much grown it's unanswerable as posed. It touches on problem-solving (uniformed+informed search strategies), decision-making (imperfect real-time decision making, alpha-beta pruning, optimal decisions, partial observability, constraint satisfaction, first-order and predicate logic, planning (esp. in nondeterministic domains), knowledge representation, Bayesian probability in reasoning, and we haven't even gotten to learning yet...
    – stoicfury
    Commented Aug 14, 2011 at 1:15
  • 1
    This question does not strike me as very good, but Cody's answer is very good. I would encourage you to consider editing your question, but I do not vote to close because I like the answer it got too much. Commented Aug 14, 2011 at 1:30
  • 1
    Any chance we could persuade you to clean this up a little bit? It's extremely casual and a bit rambling. The title-question is also somewhat unclear.
    – Joseph Weissman
    Commented Aug 14, 2011 at 3:04

1 Answer 1


No, the scientific method is still the major way of assessing truth. It may not be as rigorously applied in the social sciences (the "soft" sciences, like philosophy, sociology, history, etc.) as it is in the natural sciences ("hard" sciences, like mathematics, physics, chemistry, etc.), but the underlying methodologies and motivations are all still there and all still intact.

Perhaps you're wondering about logical proofs? Philosophical arguments are commonly made in this format, with the most common example being a syllogism. A syllogism presents two premises (or claims), and argues that the proposition (conclusion) must be logically inferred from those two premises. In other words, if you accept the premises, the conclusion must logically follow. For example:

Premise One: Tom Jones asked a question on Philosophy—Stack Exchange.
Premise Two: People who ask questions on Philosophy—Stack Exchange are smart and well-informed.
Conclusion: [Thus,] Tom Jones is smart and well-informed.

There are lots of other ways in which to frame a logical argument, ranging from the very simple to the very complex. At their highest level, logical proofs can look a lot like formal mathematical proofs because, well, they are the same. When reading standard philosophical treatises, you'll often find these arguments presented in words rather than in the more complex symbolic notation you might see in a formal logic textbook, but the methods and the process are all still the same.

The above example is actually an application of deductive reasoning, which can be more generally defined as reasoning from the general to the specific. Using such a logical framework, you make two (or more) claims about the world, and then attempt to show that a conclusion necessarily follows (or is logically implied) from those two premises. However, it's worth noting that deductive arguments are classified as either valid or invalid, not true or false.

The scientific method, by contrast, tends to follow the other prevailing method of reasoning, called inductive reasoning. Inductive reasoning can be somewhat loosely defined as the reverse of deductive reasoning: reasoning from the specific to the general. Using the scientific method, one makes certain, specific observations about phenomena in the surrounding world, and then attempts to draw a more general conclusion based on those data. In this way, unlike deductive reasoning, it does work in search of truth, and an argument framed in this way can be dismissed as false (or untrue) given sufficient contradictory data.

Finally, you should note that not all schools of philosophical thought have the pursuit of "truth" as their ultimate goal; in fact, such a goal lies in direct contradiction with many significant branches that would contest the very notion of "truth". They might argue that "truth" is simply unknowable, or they might argue that it is socially determined and therefore impossible to discuss objectively.

  • Ok, but premises always have a degree of truth; probabilities can be assigned to them. This complicates things. How can you decide on the best course of action without doing mathematics in your head or guesstimating? Or, if guesstimating is the best you can do, how do you do it right?
    – Tom Jones
    Commented Aug 13, 2011 at 9:46
  • 1
    @Tom: How do you decide on the best action? By weighing the possible outcomes. Ethics is a very specific branch of philosophy that generally follows a much less theoretical set of rules. Whether you're a deontologist or a consequentialist, you can still evaluate and compare the various options that are available. You're not searching for "truth" when you're trying to decide on an action. There isn't an infinite set of possibilities available. You can only choose the best option out of all of those that you do have available. Searching for truth here is quite futile. Commented Aug 13, 2011 at 9:49
  • 1
    I know my question is vague. Thanks. You answer is a good review.
    – Tom Jones
    Commented Aug 13, 2011 at 10:50

Not the answer you're looking for? Browse other questions tagged .