In lectures on free will, often a dichotomy between determinism and random is alluded. This dichotomy always is not a true dichotomy, there are some known and even trivial examples of non-random indeterminisms in specific contexts, which are not physicalist reductionism.

Some articles discuss the fine print of definitions, such as:

Some examples for non-random indeterminisms:

  • Divine intervention in an otherwise deterministic universe could be non-random, non-deterministic (depending on whether given gods themselves are deterministic/random).
  • Same for actions done by human souls in the religious sense, or dualist views about a strongly emerging mind (weak emergence maintains determinism/random of physical matter)
  • Non reductionist strong "downward causation". Which is non-reductionist, duh, and I am not sure if it's significantly different from strong emergence.
  • At high level of abstraction, insurance agents might distinguish damage done by machine as caused by a construction flaw (deterministic), caused by a freak accident like lightning strike (random), or the result of foul play (human choice). But those would reduce to the dichotomy in physicalist reductionism.
  • To a human observer of the universe, processes might be impossible to measure exactly in practice, making it impossible to decide whether results were determined or random, requiring ternary logic for reasoning, breaking the dichotomy for practical matters
  • In mathematic theory of random, observations of indeterministic processes might still not qualify as random given one precise definition of random or another. Those are mere technicalities interesting to mathematicians only.
  • time travel (to the past) causes all kinds of issues with definitions and timelines
  • supernatural true prophecy, predeterminism and fatalism all cause no end of issues

I am not interested in answers about such contexts, so I mention them here. I like to stay strictly physicalist and reductionist for this question.

So if we can pretend that we have a huge lab with all tools that we want, and boxes with all non-living ingredients taken from nature, metals, liquids, gases, rare earths. We might also have some simple pre-made devices, like thermometers, batteries, cables, transistors, pipes, valves, whatever you like. We can also pretend that some of those materials or devices have truly random output, like a quantum random source (everyones favorite source of true random even though the jury's still out). Is there a way we know we can assemble any device at all that exhibits properties that do not reduce to either random or deterministic events?

Can we conceive any other new base material property that could be used to create such a device?

Else to allow for the famous "third choice" in the determinism vs. random dichotomy, it would follow that we have to abandon either reductionism or physicalism, meaning assuming against Occam's razor the presence of forces in nature that have never been scientifically observed.

I am aware similar questions have been asked before, like:

But it seems recently people start debating much more loosely on this site using a third alternative to determinism and random liberally as if it came at no costs and did not require scientific scrutiny.

  • Why are Norton's dome or space invaders (from your link) not physicalist reductionism? Is it because they imply an observer, which doesn't have full information? How will you avoid an observer-based description? A god's-eye view?
    – J Kusin
    Apr 25 at 3:24
  • Both examples are technical loopholes only of applying classical physics maths to the universe, they do not describe scientific observation. A model of the universe can be described without an observers point of view, such models serve to build devices in engineering.
    – tkruse
    Apr 25 at 3:47
  • Also Norton's dome is not non-random, if such a phenomenon were to happen as described, it would be truly random in reductionism.
    – tkruse
    Apr 25 at 5:06
  • You want to say something like, we have clear proof of determinism and possibly randomness at the ontological level (physicalism+Newtonian/GR+QM). But no clear proof of this third option at the ontological level, and no clear way of creating it a la emergence, right? Well we can't fully describe most things in a reductionist manner, things we can be as sure as anything about like perception and experience. Not even the most dyed in the wool physicalist has a way. So I don't think you've levied that harsh a criticism or made a real dilemma. It's the same old reduction problem.
    – J Kusin
    Apr 25 at 5:20
  • Reductionism is just a theory of how the world works (the most popular contemporary one by most surveys). So even where cannot yet reduce some event fully due to practical limitations, in reductionism we believe that this is merely a practical matter. We impose such constrains in reductionism following Occams razor, because without it any superstition is as valid as the next. So the question is whether we already know of anything that is neither determined no random within the constraints we impose on ourselves by reductionism.
    – tkruse
    Apr 25 at 5:30

1 Answer 1


Determinism vs. random is a wrong dichotomy. That ignores free will, which is also excluded from determinism. The real dichotomy is free will vs. randomness, as they are both excluded from determinism.

  • Free will = Intentional, deliberate, purposeful, choice
  • Random = Unintentional, accidental, purposeless, chance

Can you see the symmetry? Free will, deliberate choice IS the non-random side of indeterminism.

If you should pick one card from a full deck, you have 52 options to pick from and only two optional methods for picking:

  • Either you can intentionally choose the card you like
  • Or you can pick any random card by chance

Note, that there is no third, no deterministic method to select one option out of many. If you assume both free will and randomness nonexistent, you cannot have any decks of cards.

Physical reductionism, as I understand it, assumes that there are no concepts like information or knowledge or anything else that is non-physical. Everything is physical. That naturally excludes free will, but how does it exclude randomness, too? Randomness is a purely physical thing, the very opposite of free will in every sense.

Abandoning physical reductionism does not mean accepting any "forces in nature that have never been scientifically observed". Muscle forces controlled by living beings have been scientifically observed. Even the act of scientific observation requires controlled muscle actions by the observer.


You must log in to answer this question.

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