- Since the concept of life apparently plays a role in the definition of death, we must first understand what is meant by 'life'.
- In Chapters 2 and 3, I present several of the most plausible accounts of the nature of life. I attempt to show that each of them fails. Life itself turns out to be a bit of a mystery.
- Distinction between substances and properties:
- According to the life-functional analysis of life, life involves having a property or properties.
- According to vitalism, life involves containing a substance – for Aristotle "soul", for Hans Dreisch, "vital fluid".
- V1: x is alive at t =df. there is some vital fluid in x at t
The Empirical Problem 
- Vital fluid is "imponderable" (has no weight) and invisible. It is therefore impossible to prove it exists.
The Jonah Problem 
- Suppose Jonah is inside a dead whale. According to V1, the whale is actually therefore alive, because there is vital fluid inside it (inside Jonah). Thus the definition is TOO BROAD, because it includes dead whales. (Also dead people, because there are live maggots and bacteria inside corpses3.) So simply "inside" is not enough. How about:
- V2: x is alive at t =df. there is some vital fluid inside the cells of x at t
Objection: TOO BROAD: virus could be in the cells of the dead whale
- Okay, so we need the vital fluid to be in the organism "the right way". How do we know it’s the right way? If it animates the organism:
- V3: x is alive at t =df. some vital fluid animates x at t
Objection: makes vital fluid irrelevant. The following is just as good:
- V4: x is alive at t =df. something animates x at t
Objection: "animates" is "hopelessly obscure"
The Failure of Analyticity 
- On taking a sample of an (obviously living) Martian, scientists test it and say:
- 1. The Martian is alive but he contains no vital fluid.
- This is clearly not a contradiction, because it makes sense and we can imagine it happening. (Contrast with the claim "the circle had 3 sides".) But that means that whatever "being alive" is, it isn’t necessarily containing vital fluid. And as we saw (number 3 on p. 17, chapter 1), a correct analysis of a term is necessarily true. For example, it is not possible for a bachelor to be married. That is, "The bachelor is married" is a contradiction. (This is called the failure of "analyticity" because statements whose denial involves a contradiction are called analytic truths. The fact that you can deny "being alive means containing vital fluid" without it being a contradiction, as we see from 1 above, shows that it is not an analytic truth and thus not necessary.)
DNA-ism  Suggestion drawing on Mayr’s writings:
- DNA1: x is alive at t =df. x contains some DNA or RNA at t
This is an improvement on V1 because it avoids the empirical problem. DNA is detectable because "I’ve seen it with my own eyes" – a test tube full of DNA extracted from "ground-up mouse spleens" (ugh). BUT:
Objection: TOO BROAD. The test tube contains DNA but is not alive. (This is the Jonah problem again.)
- DNA2: x is alive at t =df. DNA or RNA is contained in the cells of x at t
Objection: TOO BROAD again. A corpse4 has DNA in its cells but is not alive.
- DNA3: x is alive at t =df. x is animated at t by some DNA or RNA
Our objection to the use of "animates" in V4 was that it was "hopelessly obscure". Let’s try to fix that problem by defining animates. First attempt:
- A1: x animates y at t =df. x gives life to y at t
Objection: CIRCULAR. We’re defining "is alive" in terms of "animates", so we can’t turn around and define "animates" in terms of "life". Second attempt:
- A2: x animates y at t =df. x causes y to be able to perform the life functions at t
- EITHER this rests on life functions, and we saw in chapter 2 that nobody has been able to give an adequate account of the life functions
- OR: even if they could, it would make DNA irrelevant, because then we could define life in terms of the life functions alone:
- L1: x is alive at t =df. x is able to perform the life functions at t
(This takes us right back to the beginning of chapter 2. And we DON’T want to go there…)
- Third attempt to define "animates": treat it as a conceptual primitive – that is, something that is so basic that everybody instantly "gets it" and it defies definition (like, perhaps, "yellow").
Objection: treating "animates" as such is "tantamount to taking life itself as a primitive" which would go against the whole project of analyzing life.
- So, we’ve seen that although DNAism avoids the empirical problem for vitalism, it is vulnerable to the same Jonah problem. What about the problem of analyticity? Let’s go back to the Martian, only this time, the scientists test for DNA or RNA and report:
- 2. The Martian is alive but he contains no DNA or RNA.
- Again, this is not a contradiction (Crick even talks of the possibility of life without DNA ) so DNAism has this problem too.
Genetic Informationism 
- GI: x is alive at t =df. x contains a genetic representation of x at t
Problems for Genetic Informationism 
- TOO BROAD: Dead female frog contains still-viable eggs that can be extracted and fertilized, producing genetic representations of the dead frog.
- ALSO: living seeds in dead tomato plants5.
Footnote 1: Taken from "Feldman (Fred) - Introduction: Confronting the Reaper".
Footnote 2: Taken from "Cushing (Simon) - Fred Feldman: Confrontations with the Reaper".
Text Colour Conventions (see disclaimer)
- Blue: Text by me; © Theo Todman, 2020
- Mauve: Text by correspondent(s) or other author(s); © the author(s)