Actually your proved my point exactly because you drew a value judgement based on the argument. You inferred a value judgement based on the very principle your arguing against albeit at the extreme edge of the argument.
This is negated by me denouncing the "extreme edge" part: the set is not divided in two between humans and animals, but at least in three both above and below humans, and in practice in n. The result? The "special" position of humans is just one out of the n, and that is no longer eligible to be considered special except in a certain sophistic sense.
Note also that "humans" is not really the subset I'm using. "Humans murdering two people" and "humans murdering one people" were already different subsets above. So qualitatively, no difference of significance lies at the human/animal boundary. And this takes us further to the actually interesting stuff, for example whether all subsets of human are above all subsets of animal. Current practices vary: certain human conditions actually deprive the humans in question of rights granted to domestic animals, say.
Those opinions differing from yours?
Can be true or false. Mine is true. Chiefly because it's not an opinion, but a simple statement of fact: all human justice systems have the quality I specified, of graduating the punishment according to the crime. If somebody claims there are exceptions to that, he's wrong, barring exceptional proof.
Naturally, other systems are possible. See for example Larry Niven's classic take on why every offense should carry the capital punishment: mass murderers and jaywalkers alike are good raw material for organ banks. But mankind has never adopted such systems (or at least has not lived to document it for generations come).
What sort of philosophical musings led to the adoption of the one and only system... Can be argued about. Many such philosophies are pragmatic to the degree of being dishonest. Others are just muddy. The effect remains. Dissimilar ethical treatment is not a sign of fundamentally inequal ethics, but of a system of balances utilizing inequal treatment to achieve equality. Treating ants and humans the same, or differently, or both, is of no fundamental interest in whether current human structures for enforcing ethics stand or fall. It's just details on the practical execution of a system on a continuum of ethical subjects.
Because you used the phrase "random walk", and without any modification, that refers to a uniform random walk, as in at every X_t the PMF for X_{t+1} is uniform over the set of all possible transition states. Being a combinatoricist by education, I assumed you meant the literal object.

Apologies for misunderstanding.
You are the one owed an apology, and a clarification. I did assume that uniform random walk would inherently exist as a driving force in a system with certain basic homogeneous level of... instabilities, for lack of a more educated guess for a word. That driving force would then meet the bumpy walls and sharp corners of the computing environment, and soon homogeneity would be lost and the simplistic early terminology would grow increasingly inaccurate.
Timo Saloniemi