What is life? This question has been asked by many, from philosophers to molecular biologists, and is surprisingly difficult to define. There are some standard definitions that work well for day-to-day and practical scientific purposes. But once in a while, I wonder if there could be more. Whilst reading Siddhartha Mukherjee’s “The Song of the Cell”, I had a random string of thoughts which led me to some interesting ideas. I wrote it down and decided to make it into a short blog post to share this fun – and potentially philosophically relevant – argument with you.
Here’s how Mukherjee defined life. “To be living, an organism must have the capacity to reproduce, to grow, to metabolize, to adapt to stimuli, to maintain its internal milleu. Complex organisms have the… emergent properties… mechanisms to defend themselves … organs with specific functions… and even sentience and cognition.” He goes on to ponder asking an astrological being: “"Do you have cells?"… it is difficult to imagine life without cells.”
Could we imagine life without cells? I can certainly imagine it. The emergent properties that things with life have, including mechanisms to defends themselves and organs, could be implemented in other mediums. We are implementing incomplete parts-of-life using plastic and metals already, in synthetic organs.
Then, he goes onto the standard example: "…viruses are inert, lifeless, without cells". When I first learnt that viruses weren't considered "alive", this puzzled me. I now know that this is based on one of the standard definitions, where something alive must have the ability to maintain itself (amongst other properties) – a property viruses don't have. They need hosts, like ourselves.
What if you have an amalgamation-of-parts (AOP), like virus, that is conscious? Let's ignore the details of what sort of system is required for consciousness, and indulge me in a thought experiment. If an AOP had the ingredients for consciousness, that it could experience, express itself, but had to be tied to another organism, would it be considered alive? It cannot maintain itself and requires a constant external energy source. But if it were created – say, in our bodies – and gained consciousness as it developed, this could be considered a conscious-without-life being.
You may now think, that's a weird conclusion! Clearly, the AOP is alive! It just requires a host. By this argument, a virus is alive. I am sure I am not the first to argue this, but my interests in (weakly) emergent properties in biology and mechanisms (Craver, 2007; Mok & Love, 2023, Mok, in-theory-forthcoming-opinion-paper) made me think this is particularly interesting. The combination of parts, organized in a certain way, can be a living organism, even if not self-sufficient.
A virus is alive, as long as it has a host like ourselves. But might it be more correct – though maybe more horrific – to say that we share life?
Inanimate consciousness
Up till now, this was an interesting thought with some – I believe – valid reasoning. I wondered if I should go on, as it gets a bit nutty and fluffy from hereon, but what the point of a blog if it's just all serious ideas?
Many have asked the question of whether silicon-based life is possible. More relevant to cognitive scientists, whether silicon-based consciousness is possible. I used to think it's not possible for various reasons, and one of them is the problem of creating and implementing a biological system with "life" in silicon. However, the idea above raises the possibility of a combination of biological organism and silicon, where the consciousness may not even be a single combined consciousness, but rather a separate one in the silicon, independent from the hosting life, whilst ever reliant on it. This by no means speaks to the hard problem of consciousness. The point is simply that, if we extend the definition of life to typically considered inanimate things like viruses, we can logically extend consciousness to those same things – what one might call “inanimate consciousness”.
An alternative view – artificial consciousness without defining life
Up till this point, I had my philosophy hat on. So let me put my scientist hat back on for a moment. In my day job, I try to figure out how humans are able to do all these clever things through studying our brains. This is done by identifying some cognitive ability I find interesting (a behavior or capacity to do intelligent things) and decomposing it into parts – abstract cognitive functions or neural mechanisms – that work together explain the cognitive ability.
Why is “life” so difficult to define, and perhaps more importantly, why don’t scientists bother to define it? Probably because it doesn’t matter – its definition is irrelevant to their research questions. As scientists, we are interested in some phenomenon at hand, whether it is the cell’s ability to divide or grow, or our ability to solve a crossword or play Go, we want to figure out how it works. Determining whether something is alive or not doesn’t help us.
How about consciousness? You might think that defining it doesn’t matter too. But in fact, it matters a great deal to the cognitive scientist interested in consciousness, and relevant for our thoughts about artificial consciousness. One approach to the science of consciousness is to figure out what kind of cognitive functions require consciousness and which don’t. For example, patients with Blindsight report not being able to (consciously) see in parts in the visual field due to lesions in their visual cortex, but if you show them stimuli in their blind field and ask them to guess if there was a stimulus, they are much more accurate than you’d expect (i.e., more than if you closed your eyes and made the same judgment). Through many clever ways cognitive scientists have devised and continue to devise, it seems possible to figure out which kind of functions require consciousness and which do not (i.e., it can be defined functionally). Indeed, this is how many scientists of consciousness work on this problem.
This means that that artificial consciousness and the scientific study of clearly is possible – in brains, bodies, and even robots – whether we consider them “dead” or “alive”.
