From what I can tell, it seems that the qualifiers for something to be considered living are based on biological criteria, with the limits of observation, being entirely based on spatial order of magnitude,a subject of observation presents itself to be governing it's own motion and behavior with it's environment, and has the implied purpose of reproducing others of it's likeness. So the first part of my question regarding is something alive or dead, is how do we as what we have self declared to be living things, consider our determination of being alive or dead when observing something surrounding us to be an empirically certain conclusion?

If you want suggest well, our definition is trivial and simply for reference sake, doesn't that exclude our observations that are based on experiments we do, that are limited by our senses?

And with regard to the sentience of a subject of observation, we would consider a spider that writes messages on it's web for and equally magical talking pig to read to both be sentient, yet how do we cast the judgement that an ordinary every day spider making a web to be devoid of sentience?

The demonstrated ability in applied mathematics of the average arachnid is exceeding superior to the demonstrated ability of such a thing by the average human being, so we may as well throw out first order logic or any mathematical basis to declare something to be sentient, otherwise, everything we don't consciously decide has to be excluded as evidence of sentience, like my my motivations for deciding to write this question in the first place for example.

Let me combine both parts of the question into one example, suppose you, as a self declared living sentient being, have an indefinite life time, and are observing an inanimate lead box.

How do you know that you will not at some point in the future, be deprived of the capacity to observe this box?

And if you concede that this is impossible for you to know, then you must concede that at a time after which you cease observing the box,it is just as likely as it is to remain the way you had observed, to unfold and awaken something much like yourself, capable of observation and supposed free will.

  • "The demonstrated ability in applied mathematics of the average arachnid is exceeding superior" What makes you think spiders use mathematics when they spin their webs? Do infant children need to use mechanics before they can learn to walk?
    – D. Halsey
    Commented Nov 28, 2022 at 22:29
  • not on a conscious level no, but indeed some kind of learning process occurs out of necessity Commented Nov 29, 2022 at 10:41
  • They most definitely are using mathematics to make their, web, so my question is more do we consider something sentient only when it is capable of communicating an understanding of the abilities it has demonstrated? Commented Nov 29, 2022 at 10:47
  • 1
    If they don't use mathematics on a conscious level, I don't think you can say they use mathematics at all. You yourself might use math in describing what they have done , but that's not the same thing.
    – D. Halsey
    Commented Nov 29, 2022 at 12:58
  • Spiders made webs instinctively, so web making is not a sign of sentience. They figure out where to make them, and learn how to make better webs, the learning is a sign of sentience. Instinctive web making is not based on knowing mathematics. Neither is learning how to make better ones.
    – Dcleve
    Commented Nov 29, 2022 at 16:19

1 Answer 1


You appear to be looking for definitive analytic criteria. This is not available, as spelled out in the answers to this prior closely related question: Philosophy and the question 'When is a robot considered alive and thinking?'

As the answers all note, there IS no agreed explicit criteria for either life, or sentience.

This is symptomatic of the general issue of how to do empiricism, and the ongoing effort by the analytic tradition in philosophy to fit empiricism into an analytic framework. Kant, in The Critique of Pure Reason, spelled out how our universe is contingent, and CANNOT be characterized analytically. The disentanglement of empiricism from analytics has slowly been progressing further since.

A major step was when mathematics was shown not to be analytically "necessary", when non Euclidean geometries were discovered, and then worse, when our universe was shown to be non-Euclidean. Math is now considered to be an infinite set of potential systems, and part of empiricism is to determine if one of them seems to fit an aspect of our universe. Note "seems" is key here -- we never can be certain whether a math fits our universe or not.

Popper's approach to science emphasizes this embrace of uncertainty, and that all of science is tentative. Note that in Popper's model of science, which is to do a) exploration of a subject area, b) then directed exploration to study interesting features, c) development of speculations that one then explores further, d) hypotheses that one then tests, then revise based on testing, and e) theory development and testing -- definitions are one of the LAST things we end up with, basically only when hypotheses are solid enough to begin turning into theories. Analyticity needs definitions to be nailed down, and science does not get close to there until the end of step d.

Quantifying science confidence has been an ongoing goal even of Popperian science, but one with a very unsuccessful history. Popper was challenged in his assumption that science is gradually getting better insights about the world by Kuhn, and Popper proposed a method to measure this "versimilitude". His standard, and ANY standard for versimilitude, were shown to be analytically invalid. https://en.wikipedia.org/wiki/Verisimilitude.

Popper's falsification was also challenged, and is best refinement is Lakatos' model of Research Programmes. https://1library.net/article/lakatos-conception-research-programs-lenoir-thesis.yr64rwjy While Lakatos' thinking is seen as a "better" (IE more verisimilitude -- LOL) match to scientific practice than either Popper or Kuhn, Lakatos then faced a challenge of characterizing when a Programme is Regressive vs. Progressive. He proposed a method of measuring this, but his methodology faced the same fate of being analytically invalid in some limiting cases as Popper's measure of Verisimilitude.

A similar fate has been encountered in scientific measures of statistical confidence. Critics of frequentist statistical methodologies point out that statistics uses a lot of "judgement" criteria that cannot be justified analytically. Bayesian statistics seeks to correct the subjective "judgement" aspect of frequentist statistics, BUT both the "prior" probabilities, and the applicability of the new data in changing degree of confidence involve judgement calls in Bayesian statistics as well.

A general principle one can draw from the history of repeated failures of analytics to characterize empiricism analytically, is that empiricism is NOT an analytically justifiable activity -- it is instead justified pragmatically. And pragmatism ultimately references subjective judgement, rather than reason, as the foundation for truth.

So can we spell out a set of standards and degrees of confidence whether something is living, or sentient? Yes, we can. BUT -- those standards themselves are not justified analytically, but as pragmatic judgements. And for both life and sentience, we don't have anything close to an expert consensus to even reference pragmatically.

  • It's not that I was seeking an answer that's deterministic or analytical, i knew that to not exist, just an answer with a well informed opinion with regards to the subject of the question. I don't think such questions deserve to be closed. I realize that it is against stack exchange policy to permit the exchange of OPINION, but applying this restriction is a little dubious, especially for philosophy Commented Dec 2, 2022 at 11:14
  • @AdamLedger There are various proposed pragmatic models. One answer on life listed the Biology definition -- with its set of criteria. That is useful. Another useful one is that Freeman Dyson proposed is purely functional -- Organized systems that use entropy to create information. Neither capture the full concept, but both are useful. For sentience -- one approximation is to infer by analogy that any adult mammal will have a similar enough neural stricter to reasonably infer sentience. A functional approach is to look at apparent agency, or at learning, and both include even bacteria.
    – Dcleve
    Commented Dec 2, 2022 at 23:33
  • Well that addition certainly strikes a chord in me, thank you for taking the time to converse with me on such a "touchy" subject. Particularly the part regarding organized systems using entropy to gather more information. For a very personal example, I have a problem with alcohol. So I decided that only on my birthday I will allow myself to completely indulge this addiction, and the night commences with me buying an $100 bottle of french absinthe, 24 hours later I woke with at least one broken rib this year, more later I'm still in pain. Commented Dec 6, 2022 at 17:03
  • I guess my point is, I came to a point that I knew subconsciously that disassociating myself like this is pointless and harmful, but I never really reconciled this with any real conscious agency, So something in me decide to limit such an activity to once per year, and it was clearly just as dangerous as it always was, and again very little memory of what happened exists, but the painful physiological reminder, coupled with my decision to only do such a thing on my birthday, has left enough of an imprint to think about never drinking to such excess. Commented Dec 6, 2022 at 17:10
  • I just feel like perhaps at least sentience, is something far less tangible than we allow ourselves to consider, It's human nature to consider the universe revolves around us, rather than us listlessly orbiting within it Commented Dec 6, 2022 at 17:13

You must log in to answer this question.

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