Here's an argument that I've heard a number of times from friends and on the Internet:

"The ratio of what we know about the universe to what we have yet to discover is so small - it is therefore illogical to draw conclusions from the crumb of knowledge - how can we be sure about any proposition concerning our universe which is based on the limited and insignifiant observations of mankind."

In this case the speaker is appealing to our modesty in order to make us willing to accept the possibility that gods, gosts, ufos, etc might actually exist in our universe.

Does this well-trodden argument have a name? It seems like an appeal to lack of evidence to the contrary, but it's slightly more since were we to accept it it seems to undermine the value all propositions based on evidence.

Are there any notable refutations to this argument?

My attempt to refute it was to point out that it appears to be a straw-man of human discovery. We rarely make final pronouncements, we only say which theories are best supported by the evidence available right now.

  • Possibly related philosophy.stackexchange.com/questions/427/…
    – Joseph Weissman
    Commented Feb 20, 2012 at 18:17
  • If he's saying that we know little about the universe, it wouldn't make sense to understand the speaker as trying to make us willing to accept the possibility that gods, ghosts, ufos, etc might actually exist in our universe. Unless you're omiting something of the integral text.
    – Red Banana
    Commented Apr 13, 2013 at 23:15
  • If we know nothing how do we know that we do indeed know nothing? Sounds like self contradictory statement.
    – Neil Meyer
    Commented Jun 3, 2013 at 19:01

6 Answers 6


Yes, appeal to lack of evidence to the contrary has a name -> proof of burden.

If they are using it in a super being / religious way, throw Russell's teapot at them. Of course we can not know 100% if a super being exists, the same way we can't know if mute, transparent, ghostly miniature unicorns don't exist - but it is the believers burden to prove it - not the disbelievers.


That we can very reliably make predictions on the basis of what we do know (or, rather that in the past we have been able to) is the best counterargument I know of against that argument.

Although this appears in various guises everywhere from Popper to coherentism, the simple observation that we routinely do not walk into walls (and manage to build fairly solid ones) shows us that we have a keen grasp of a good number of phenomena on our temporal and spatial scale. We thus have something of a bound on the depths of our confusion: at least our universe has to be consistent, somehow, with the regularity we have managed to observe locally. (One should still be cautious about being too confident of proclamations about the universe, but with the can't-know-anything genie back in its bottle, we can at least entertain the notion of trying to know more than nothing.)


I think that we have to turn to the great philosopher Rumsfeld, who famously opined about "known knowns", "known unknowns", and "unknown unknowns."

The size of what we don't know about the universe is an unknown unknown; we necessarily have no way of knowing how much (or how little) there is we don't know.

So: all the more reason to examine rigorously and build upon that which we do know. The fact that our knowledge may be comparatively small makes it all the more precious; the argument from ignorance is self-refuting.

  • 1
    We have a lower bound on how much there is we don't know: The known unknowns are part of the unknowns, therefore the latter are at least as many as the former. Especially, should we find that there are far more known unknowns than known knowns, then we can conclude (under the assumption that there are no unknown knowns) that the total number of unknowns is much larger than the total number of knowns.
    – celtschk
    Commented Oct 19, 2012 at 20:16

I can not refute the argument as it is perfectly sound.

How can a frog at the bottom of a well know about the vastness of the oceans?

In defending science, we rush too far ahead of reality. Science has helped us in the pursuit of knowledge, but in relation to what we yet do not know, we have discovered a tiny amount and therefore everything we think we know may well get overturned some day, or indeed we may never discover just how much we misunderstood reality.

This has happened time after time in history. We think we know something, then we discover we don't.

  • 2
    The Zhuangzi reference is cool but the leap from that to "In defending science, we rush too far ahead of reality" is confusing to me. It's not clear what you're suggesting there, despite the fact that the remainder of the paragraph is quite clear—about continually rewriting/rebuilding knowledge. I don't see how such a notion relates to us "rushing too far ahead of reality". I feel like rewriting/rebuilding knowledge is our way of getting ourselves back to reality... Could you elaborate on what you meant there?
    – stoicfury
    Commented Feb 23, 2012 at 20:14
  • There's a large body of knowledge we can be very confident to never be found wrong; at worst it will be found approximate. Just as quantum mechanics has not invalidated classical mechanics (for example the trajectories of the planets are as accurately described by the classical equations as ever, despite quantum mechanics telling us that strictly speaking they don't even have trajectories), we can be completely confident that the laws of quantum mechanics will continue to hold in the regime where we have tested them, even if we one day find out that they don't hold at the fundamental level.
    – celtschk
    Commented Oct 19, 2012 at 20:24

I think your "no final pronouncements" idea exists in the philosophical literature as a form of Anti-realism about Science - not only do we not make final pronouncements, but this abstinence is based on principled grounds concerning insurmountable barriers to any such finality. We can adopt an inductive Pessimism concerning our ability to be totally correct about the consequeces of our predictions, and thereby consider that there will never be a point in which we can take ourselves to be completely right.

To prevent your opponent thinking of this as a victory, you can respond by noting that rather than our knowledge being in question, it is the clarity of the notion of "the universe" considered in the argument that is now in need of evaluation. The common-sense view does not come without potential difficulty, as discussed in the SEP article concerning Challenges to Metaphysical Realism. And in either case, we will have either explained how we can, in fact, talk about things as they are, or proposed a view of factiveness that does not depend on the totality of "the universe" to account for knowledge.


The ratio of what we know about the universe to what we have yet to discover

Here, a person who is arguing for the insignificance of the totality of human knowledge is presuming to know the quantity of the total gross sum of knowable knowledge, and then calculate from that... a ratio. He does this, presumably, to point out the limitations of other people's knowledge. But he seems to over-estimate his own knowledge right from the outset. And here is why:

how can we be sure about any proposition concerning our universe which is based on the limited and insignifiant observations of mankind.

Here, he attempts to link knowledge to some presumed total knowledge of the total universe. The argument, as I have heard it stated, is that since all things are related, how can we know the facts of one part of the universe without knowing how it is being affected by other parts. I am trying to express the argument in it's most general form but here are a couple of examples:

  1. How can we know the specific conditions of a solid object when
    fluctuations are eminating from supernova we don't yet know about
    which may affect our measurements?
  2. Why should we believe some theorem today when it will be refined or overturned in ten or 100 years?

The problem with both of these scenarios is that they presume the existence of some grand total sum of universal knowledge. Item 1 presumes that a measurement we take today will be so trusted as not to require error checking. Item 2 makes a common, but only slightly more complicated error. It makes the mistake of believing that theorems like those of Newtonian physics were meant to apply to all future ranges of experience which were to be discovered. They were not. Newton formed his theorems on the scale of physical properties of which he was aware at the time he wrote his theorems. His theorems still work and are used within that range today. They were not overturned and eliminated by the discovery of a new range of physical properties. Furthermore, all those theorems that are found to be false are only found to be false by discovering that which is true. Each time we invalidate some theorem we are saying we know this is false because of these other facts of which we are certain.

concerning our universe which is based on the limited and insignifiant observations of mankind.

Insignificant to whom? There is no universal significance. It is quite presumptuous for the asker of this question to estimate the significance of the set of human knowledge on some universal scale. How has the asker positioned himself to be the evaluator of this universal significance? He is only aware, and can only be aware, of the significance to himself and to a lesser extent, other humans, of any knowledge set. The asker looks down and sneers from his perch upon the infinite knowledge set at how small and insignificant human knowledge is.

But there is no infinite knowledge set. It would require a computer the size of the universe moving at the speed of light and containing all the human minds and their memories in order to know their thoughts. It is a ridiculously comprehensive set of data.

And this is what the asker wants to use as a yard-stick to measure human knowledge against.

I don't know if this question or argument has a name yet, but I think it derives from neo-Platonism. The neo-Platonists believe in infinite knowledge which can be achieved through divine insight. It derives from Plato. This is only a rough estimate as to the source of this question.

You must log in to answer this question.

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