Every school biology student is taught a definition of life, generally along the lines of it being a self-sustaining, reproducing chemical reaction--or something like that but elaborated in one way or another. Perhaps after Darwin we might have to add reproduces and evolves adaptively. But one view is that being biologists or not, experts or not, we really do not actually know what life is.
An interesting commentary in the NYTimes argues that we do not have a really adequate definition of life, even now after two or more centuries of modern science. A crystal is regular, organized, and self-reproducing, for example, surely characteristics of life. A virus does not, by itself, reproduce. The author, Ferris Jabr, is a science writer, not a working scientist, but the commentary is knowledgeable from an outsider's point of view.
In the confusion or lack of agreement or whatever, Jabr suggests that 'life is a concept, not a reality'. Is this a reasonable way to view things? I think not, even though the points made are all reasonable and recognizable--and they do relate to important issues that we face in modern genetics, biology and evolution.
Jabr's view borders on being Platonic, but at the same time of being very perceptive. Platonic 'ideals' existed somehow, but were immaterial. All we actually see or experience are imperfect instances of an ideal. We see dogs and tables, and we recognize them (and can tell them apart), but yet we do not see 'the' dog or 'the' table. Plato argued that there are such entities, but they're not real things in this world. Yet, we can operate perfectly well without experience the real thing, so to speak.
Of the classics, Aristotle had a view more like what we would say today. He did not accept the reality of Platonic ideals, and was more of an empiricist, that the dogs we see are the dogs that are.
The human genome sequence doesn't exist!
We do something similar, but it is not Platonic and it can be misleading. For example, we all the time hear people, including scientists, talk about 'the' human genome, and we often speak people as having 'copies' of 'the' human genome. The same applies to any other species genome. But in fact there is no such thing as 'the' human genome! It is treated as a Platonic ideal. But it is an arbitrary reference sequence that no one person has ever had. It is a very very useful tool as a reference, because it gives us a way to find corresponding places among people in their instances of human genomes. But it is not a Platonic ideal, and there is no such thing.
We each have two instances of human genomes, but they are not copies of the reference sequence, and not even of our parents' genome instances, because there are always mutations. And for the same reason, all our cells have instances of human genomes that are not 'copies', in a sense not even 'imperfect copies' because there are recombinations as well as mutations and other changes.
Not only this, 'the' human genome is not a 'normal' genome. The donor(s) whose sequence(s) was/were assembled to construct the reference had characteristics that are not given for reasons of anonymity and even if 'perfectly normal' at the time their blood was taken, sooner or later they will get sick from something that may be genetic.
We indulge in similar things all the time in biology. We have type specimens or nowadays reference genomes for 'the' lion, mouse, horse, and E. coli bacterium, etc. You can look up 'the' sequence for the Apolipoprotein E, Cystic Fibrosis, or Hox3A genes, but they, too, are references or sequenced instances.
There's no such thing as a 'normal' genome, either--not figuratively, not literally!
One might say that, well, yes it's true that 'the' human genome doesn't exist, but that we can use the idea of a normal genome, and even identify its sequence. That sounds obviously right, but it is not possible. The concept of 'normal' arose when nations began collecting population data. Then, one could take an average of one or more variables, and use that to represent the norm or, slipping a bit erroneously in verbiage, the 'normal' person was. He might be 160 lbs and 42 years old, and so on. If you use many variables, no such actual normal person exists: the average values comprise a Platonic ideal of the very idea of the norm, or of 'normal'.
Back to the question
This is relevant to the question whether life is more than a concept but, perhaps, less than a reality. We work happily away in our labs or in the field every day with our quasi-Platonic concepts. Much of the time, it doesn't really matter.
Ecologists can study predators and prey, or immunologists can study infectious disease, without worrying about some of these details. That's because, while we often use the 'the' rhetoric we know that we're dealing with collections of variation around some kind of central tendency. There may be exceptions, but they are rare: We can tell a human or a dog when we see one so to speak.
There are, however, important times when our definitions and our Platonic idealism do impinge on valid science. If you do an experiment in 'the' mouse, or even 'the C57/Bl6' mouse, you are idealizing unless you are very forthright about not only the variation you see among instances of C57's, or that these represent any other mice, much less, say, humans. If we keep the approximate, variable instance viewpoint in mind--often hard to do in the press of high-throughput research--then we can make progress and do so every day. The danger is oversimplifying by assuming that one's data represents an ideal with universal applicability, and so on. We do see that sort of thing every day.
Thus, some of the work we write about in human biomedical genomics, or in constructing evolutionary stories, we often find people clearly but inadvertently idealizing in Platonic ways. Its inadvertent, we think, because they know when asked specifically that the specificity of the assertions is often not justified or has unclear precision. But the haste of today's science too often leads to these 'lapses'.
Still, that's a caveat and overall we function quite well in an Aristotelian world. We don't usually need to know what life 'is' to do our work. But the deeper philosophical question doesn't really explicitly arise, except, of course, for those trying to infer the chemical origin of life, or perhaps those hyping about life on other planets who are essentially defining 'life' as what we know about here on earth.
This doesn't vitiate the philosophical interest raised by Jabr's thoughts. But, like asking what is 'existence' or what is 'probability' or what is 'mind', we can enjoy the mysterious musings we can indulge in over those questions. They just rarely directly effect the study of life. Like pornography according to Justice Potter Stuart, we know it when we see it.
At the same time, as we note above, the haste for quick answers can lead to erroneous and even harmful lapses into Platonism.