In artificial biology, a distinction must be made between 1) genetically modified organisms, which are indeed living organisms, even if they are enslaved, 2) synthetic biology, which is limited to synthetically reconstructing a given genome, which has succeeded in effectively recreating a living bacterium (perhaps a mammoth one day), 3) finally, the project of artificial life, i.e., the creation of a living cell from scratch using a minimal genome, or even bases other than DNA or RNA (such as APN).
We are nowhere near this yet, but what is interesting is that it raises the question of the creation of a new form of life, because we can be fairly sure that this artificial life will have nothing to do with real life. Indeed, life is evolution, whereas artificial life must not evolve, or only marginally, in order to meet our technical requirements. Rather than living organisms, what we would end up producing would be biological machines, possibly programmable. Nothing truly alive, because life cannot be reduced to reproduction or metabolism; it is plasticity, a process of transformation through interaction with its environment. Life without evolution is like intelligence incapable of learning: a contradiction in terms. There is no life cut off from its origins, without a history that it continues (genetic heritage) or without a world that it inhabits and which constitutes it (diachrony and synchrony).
Life can be defined in different ways, none of which is entirely satisfactory. According to NASA, any system capable of self-maintenance and reproduction by manufacturing its own constituents is alive. Another common definition defines life by its flows of energy, matter, and information. This is the systemic point of view, but it also misses the essence of what makes life alive: its capacity for adaptation and evolution.
Until now, I have favored the definition of life as reproduction, because it is reproduction that triggers the process of selective chemistry that leads to the internalization of externality through Darwinian evolution, selection by result constituting a downward causality whose memory is retained by the genome and which will be enriched and complexified by future generations (which is why life always comes from life). However, reflection on artificial life shows that reproduction is not a sufficient criterion and that variation, adaptability, and evolution are essential characteristics both in the functioning of the cell and in animal activity. One might even consider that this variability and the ability to cope with the unexpected constitute the subjectivity of living beings, their autonomy, their interiority, but also everything that artificial life would want to get rid of. It is not clear what purpose evolutionary artificial bacteria would serve that natural bacteria could not serve just as well, even if diversity generators could be used in them as detectors, for example, but this is not the same thing as constitutive variation. We can therefore consider that life cut off from its vital force is not alive but merely a biochemical automaton.
Jean-Jacques Kupiec (and Miroslav Radman) emphasize the random and changing nature of genetic expression, which is not a linear program but involves random variations and inextricable combinations of multiple proteins subject to Darwinian processes, requiring systemic concepts beyond any particular gene, and taking into account the constraints that act on it (compartmentalization in particular).
Miroslav Radman adds that the internal mechanisms of diversity generation (GoD) seem to prepare the cell for any future eventuality in a race against external threats that do not yet exist (because it must be able to respond immediately to a new virus or a new environment).
For his part, John Stewart emphasizes the non-separability of the organism and its environment in this adaptive and autopoietic process between the organism and external reality, just as one cannot separate the “building blocks” of life any more than one can separate the stones of a bridge.
We might add that an organism cannot be separated from its viruses, which are vectors of information between organisms and serve to regulate the species according to population concentration, thus forming a multicellular superorganism made up of the viruses and bacteria specifically attached to it (so the further we move away from the ocean's surface, the more viruses there are regulating the bacteria).
Much like the question of the speed of light seemed a very minor issue before becoming central, it appears that these singularities of life are not minor details but rather the very essence of life, giving it the internal dynamism that distinguishes it from cellular automata, what makes life exceed life and open up to the world it explores and against which it bumps as it moves forward through trial and error, betting on the future and persisting in being through constant learning and the capacity for irritation that drives it not to give in, to innovate and adapt in interaction with others.
Besides this, artificial life will be nothing more than a soulless cell (devoid of any subjectivity), with just a membrane, a minimal metabolism, a ribosome, and a few genes to constantly produce a few useful proteins. The minimum still includes reproduction, without which artificial life is useless. It must be proven that this is possible, because it will not be possible to completely imitate living cells and their natural variability, nor to eliminate inevitable errors. It will therefore probably have to be done in a completely different way.
It is not certain that it will be possible to separate reproduction and innovation (or reproduction and replication, which amounts to the same thing), as Antoine Danchin is attempting to do. Furthermore, the result is likely to be less effective than genetically modified bacteria, but the attempt will certainly be instructive and harmless, since artificial life without the ability to adapt is not viable outside its culture medium.
If we want to build a reliable cell factory (who would want to fly in an airplane that innovates in its behavior?), we will have to omit the genes that enable innovation. The consequence will be that the descendants of these cells will age and then die out. The factory will have to be rebuilt periodically. But this has one obvious advantage: the associated risks, which are linked only to the possibility of innovation in life, will be reduced to zero. Antoine Danchin
What artificial life teaches us is what differentiates it from real life, which is linked not only to its environment, which we must learn to preserve, but also to its transformations, which we must anticipate. As I said in “The Improbable Miracle of Existence,” life is the improbable responding to the improbability of the world, the nostalgia for unity torn apart by the contingency of being, the hard desire to endure in an unpredictable and changing world, the resistance to periodic catastrophes through the emergence of information (genetic memory) and its anti-entropic capacities for regulation and organization.
What must be added is that life is not exhausted in the result but is essentially a continuous process of adaptation and transformation, not just reproduction (without which there is no evolution). Life does not aim for a state of satisfied ataraxia, but is instead a ceaseless activity. This is what distinguishes living beings: they are always in motion and agitated by currents, chemical reactions, construction and destruction, an agitation that is finely regulated by the final result and caught up in a general pulsation. The important thing to understand is that this agitation is not a disturbance, any more than diversification is; it is life itself, a life that is not intermittent (except in very rare cases) but continuous (albeit cyclical) and cannot be reduced to its own preservation. Contrary to a logic of identity, we are embarked with life on an adventure whose end is not predetermined and in a dialectic with our environment that constitutes us entirely in the long term while leaving us unfinished. What seemed marginal, attributed to reproductive errors, turns out to be central, with the process of interaction and adaptation to the environment constituting its excessive and invasive character, to which life owes its insistent existence.
At present, despite some spectacular results from Craig Venter, there is nothing less certain than the possibility of creating true artificial life, let alone reproducing the first stages of life, where time and the succession of events are the limiting factors, even if John Stewart is convinced that it will not take that long and does not constitute an insurmountable barrier. This is undoubtedly where computer science could accelerate the selective processes and resolve the question theoretically (through reverse engineering) before it is verified experimentally. The mystery of our origins will remain one of the things that triggered this cumulative history that we continue in our confrontation with the limits of the biosphere and that drives us to invent our own life.
To say that we will not achieve true artificial life, at least not for a long time, or that artificial life is not life, does not mean that this research will produce nothing at all. There will be results, that is certain, at least for genetically modified organisms that already exist and programming based on biological components (biobricks). More than fantasies about the creation of a new, quasi-divine life or even an improved human being, it is indeed poorly controlled GMOs that should be the focus of our attention and to which the precautionary principle should be most strictly applied. Moreover, the greatest danger here, after greedy multinationals and rogue states, is undoubtedly biohackers who engage in genetic manipulation in their kitchens...
