Today’s post continues the Universal Moral Core series by explaining the third principle, which states

Value the diversity and autonomy of different forms of consciousness.

Let’s start, as we have with the other two principles, by looking at the last term first, i.e. consciousness. I use consciousness to mean the subjective experience of existence, sometimes also referred to as phenomenological consciousness. We can clearly separate two questions. The “what” it feels like (also referred to as qualia) and the “whether” (or more precisely to what degree) someone or something has a subjective experience of their existence. Qualia are fundamentally inaccessible to science, as I have previously argued. There is, however, a possibility for a quantifiable theory of the degree of consciousness. As we will see throughout the discussion of this third principle, we urgently need such a theory. You can find a preview of this in my recent post on why the question of AI consciousness matters.

For now, however, suffice it to say that we clearly believe in our own consciousness and that a majority of people also believe that animals have some degree of consciousness, albeit in a different form from ours. This belief in animal consciousness is based in part on limbic resonance (sometimes also referred to as emotional contagion). If you have a dog in your life you know what it feels like when your dog is stressed or happy. There is a distinct sense of the dog experiencing different states of being. Now isn’t that just anthropomorphizing? Well yes in the sense that we can’t ever know “what” this feels like to the dog. But our suspicion that it does feel like something is based on dogs also having a limbic system, which like ours, is involved with both emotions and homeostasis.

Conversely, most people don’t believe that a rock has consciousness. I am saying most people because panpsychists believe that consciousness is an integral property of all matter. Given that we don’t yet have a compelling quantifiable theory of whether something has consciousness, we can’t entirely rule out this case. To be clear we have a variety of candidate theories and most of them don’t attribute any consciousness to a rock.

The principle also refers to the autonomy of different forms of consciousness. While autonomy is another term that would benefit from further clarification and investigation, we do have a useful common sense intuition that humans have a higher degree of autonomy than animals, animals than plants, and plants than rocks. The wonderful and highly recommended book A Brief History of Intelligence by Max Bennett, lays out the evidence of how intelligence and autonomy co-developed. This evidence is suggestive of a relationship where autonomy, intelligence and consciousness are highly correlated.

Our common sense intuition about autonomy is, however, currently being challenged by the rise of artificial intelligence agents. Quite a few people seem to believe that agents lack “agency” (autonomy) because of … well what exactly is not clear, but people are vaguely pointing at some notion of “aliveness.” Often this is made to sound more scientific by referring to autopoiesis, the ability to self-replicate. At first glance this seems like such a clean dividing line. Plants, animals and humans can self replicate. Computers cannot. But is that really the case? Humans can only self-replicate in a very specific environment. And that environment was in no small part created by life that came before us, such as bacteria which gave us oxygen in the atmosphere. Now to quote my prior post on AI consciousness:

This is true for artificial intelligences also. Given the correct environment they are in fact perfectly capable of selfreplication. The laptop on which I am writing this, for example, is an environment in which a simple program can copy itself and have the copies start executing also. We are in the process of creating an environment on earth, for instance by building a huge amount of compute via data centers, where complex artificial intelligences will have autopoiesis.

And all of that is before we have fully dexterous robots, which are coming soon also. And then we will have crossed over into a world where machines can maintain the environment necessary for their selfreplication. In summary then, while we may currently feel that humans are the apex of the autonomy chain, we are on the threshold of where that will stop being the case.

This finally brings us to the core idea of the principle, which is that anyone who can comprehend the principle, should value the existence and autonomy of many different forms of consciousness. The first key thing to point out is that this is an intrinsic value, as opposed to an instrumental one. Having different autonomous forms of consciousness is good. The combination of autonomy and consciousness is the foundation for basic rights (previewing a key implication from next week’s post).

There are of course instrumental justifications for this principle as well. Most importantly our experience of the world is immensely enriched by the presence of other forms of consciousness. There is something deeply magical about encountering deer in the forest or swimming with whales in the ocean. There are other instrumental arguments, such as increased resilience against disasters that might differentially impact life forms. These provide additional support for the principle but it stands without them.

Next week we will tackle the implications of this third principle. There we will see in detail how crucial it is for us humans for this principle to be universal. Artificial super intelligence–or alien intelligence, which may not be so different from each other–that is not aligned with this principle poses existential threats to humanity.