Review : Planta Sapiens

After a string of marvellous books on animal sentience, unfortunately it's time for a bad one.

If you do a search for "plants" on this blog, you'll find my position on whether plants can be said to be intelligent or self-aware* has long been on the skeptical side of uncertain. In particular, there have been experiments claiming that plants can learn and remember, that they can make choices, and in particular that they can show Pavlovian responses. That one in particular strikes me as potentially a good route to convincing me that they have inner, conscious lives. Which makes it disappointing that the experiments have problems with reproducibility, despite some criticism being unduly harsh (you might also recall the findings hinting at fungi also being capable of making intelligent, sentient choices).

* I plan to try and define these terms a bit better in a future post, but for now I'll use them very loosely.

Likewise, I think that panpsychism is a philosophically valid, intelligent position but I just don't like it very much. The idea that everything has a consciousness, however rudimentary, just seems extremely strange and largely pointless to me. I lean considerably more favourably towards the idea of "biopsychism", in which all biological life has a consciousness, but even that I don't find especially brilliant.

Still, if you can convince me that plants are conscious, I'll be very much happier about biopsychism and would give serious reconsideration to panpsychism. As I said, some of the experiments are very intriguing, so I'm definitely persuadable on this issue.

All told, I'm the prime audience for Paco Calvo's Planta Sapiens. This purports to be a scientific account of plant intelligence written by a philosopher-turned-botanist. His aim is to convince the reader that plants do indeed have minds, but I have to say that while not without some interesting points, as it stands the book is a disappointing failure that's frequently an actively irritating experience to read.

Unlike other recent reads I can keep this to one post. Of the few genuinely interesting findings that Calvo does highlight, I'm going to have check the references and give them their own post in due course. For now, I confine myself to the book.

The review bit (skip this for stuff about plants)

Calvo begins quite properly with an exhortation to keep an open mind. And he keeps going with this. And going, and going, and going, right through the entire book. All too often I could not but help thinking, "shut up and tell me about the plants !". Strangely, he also likes to detail the exact progress of his career, making odd and irrelevant references to his own and his colleagues employers and research institutes. At times the anecdotes did help me warm to the guy despite the lack of substantive content, but by the end the final chapter was just a pointless, outright stupid rant and I wanted to tell him to get over himself.

Yeah, it's not great. It's a short book, but Calvo chooses to use most of the space for talking about non-plant related things. These include his career, the need to keep an open mind, other people's careers, socialising at conferences, the need to keep an open mind, philosophy of mind, unrelated analogies, the need to keep an open mind, and the harsh responses from mainstream botanists. Oh, and of course, the need to keep an open mind. Again.

Dude, I'm open-minded enough to be reading the damn book. You don't need to force the open-mindedness down my throat. Give me something to be open-minded about, for goodness' sake. It's bordering on insulting, to be honest. 

Worse, even when plants do make an appearance, their behaviours are rarely (but importantly not never) as interesting or as convincingly conscious as Calvo seems to think. Frustratingly, his anecdotes do at least demonstrate that he's aware of why people disagree or don't think the evidence is that convincing, but he fails to address their points. He spends a lot of time explaining why plants could be conscious, a position I'm already sold on, but my guess is that no more than 5-10% of the book actually contains anything about plant behaviour which is anywhere near interesting enough to warrant this.

Now to be fair, it's not that some of the extraneous content isn't valuable; some of it is genuine rubbish but probably about two-thirds could have some value. If this is your first encounter with the world of philosophy of mind, I suspect you could do worse : his descriptions of the problems and the histories behind them are lucid and readable, and taken on their own they're very light but perfectly decent. The problem is that without the experimental plant-related backup, all of this is just little more than wishful thinking, a thick layer of icing with precious little cake. It's a bit like having a car manual that tells you in tedious detail about what a magnificent car you own but forgets to describe how to switch on the lights.

In other words, yes, plants could be conscious. I agree. NOW BLOODY SHUT UP AND TELL ME THE EVIDENCE THAT THEY ACTUALLY ARE !

I'm giving this one 4/10. There are brief moments when something really interesting crops up, and most of the superfluous content isn't actually bad in itself*; I give further credit because Calvo does duly present concerns about experimental verification and doesn't take everything on face value. It's just that the book is too short, but not in a good way. It's like a cavalry charge without infantry support : most of the content would be perfectly good as reinforcement, but is useless on its own. It's also quite repetitive, and while I value some repetition to drive a point home, this much is excessive.

* Some is though. I don't need to know the backstory to every conference or what sort of wine you were drinking. A great deal of this feels like a blog post that hasn't translated well into a book format and just becomes outright annoying.

Even so, what blog-worthy stuff does the book contain ?

The content

1) Interesting plant behaviour

A hopeless beginning

Calvo begins unexpectedly with anaesthetics. Plants succumb to the same chemicals used to induce unconsciousness in humans. Calvo starts off gently here, highlighting that more important than this is the fact that plants do have some kind of active state that can be suppressed. And this is not at all an unreasonable claim which is worth bearing in mind

A running theme that Calvo develops early on is that plants may need intelligence precisely because they're fixed to one location. Animals, he says, have the luxury of being able to correct their mistakes. True, they have to have additional brainpower because of all the extra activities they do, but they can generally afford to try again if they mess up (think of how most animal hunts end in failure). For a plant, because its movement is largely growth-based, each failure is more costly. They can't self-correct so have a greater propensity for fatal mistakes. So they may, he argues, need to think.

But this isn't at all convincing by itself. It might be advantageous to the individual, but it's hard to see any species-wide requirement for thought : plants vastly outnumber animals and are in no danger of simply being gobbled up. Far more persuasive would be examples of plant behaviour that are too complex to explain by mere stimulus response.

Calvo believes very strongly in a softly-softly approach, so he begins with a common, simple behaviour : circumnutation. When reaching for a support, plants circle around in a lasso-like motion before finally gasping on. This can be surprisingly rapid, with a complete period of about two hours in some cases, with the final contact taking only a few minutes or perhaps less (because the rest of the motion is so slow, this is difficult to capture in time lapse).

But what's supposed to be intelligent about this I've no idea. A circular motion seems like a decent and simple enough way to cover a wide area in a search for structures. There doesn't seem to be anything about it which meets Calvos' own criteria for cognition, e.g. being adaptive, flexible, and goal-directed. Nor is there anything all that remarkable in plant's producing chemicals to influence animals. And while the ability of some plants to track the Sun on cloudy days (though not very accurately) and anticipate where it will come up the next morning is quite interesting, Calvo himself describes quite convincingly how this is due to purely physical, mechanistic processes rather than mental computation.

Slightly more intriguing are that plants sense a wide variety of external stimuli : water, light, touch, chemistry (airborne odours and soil chemistry), temperature, etc. They have to in some way compare all of these different strands of information and juggle them against their competing, individual needs. So at least some of the rudimentary requirements for consciousness are there, but it doesn't follow that they necessarily actually implement this. Everything could still perfectly well be pure stimulus response with different thresholds governing the result. Likewise, while plants undoubtedly communicate with each other through chemicals (either airborne or through direct physical contact), Calvo heavily implies that this means they must be intelligent, but why this should be so isn't at all obvious to me.

A few promising developments

We do eventually get some more intriguing behaviours. Calvo mentions the Pavlovian experiments I referenced earlier and, to his great credit, is careful to stress that these results haven't been replicated. But he's more bullish about describing other, extremely similar results. Apparently both pea plants and strawberries are capable of learning that particular light levels (either bright or dark) are associated with higher nutrient levels in the soil, putting down roots in greater amounts in the areas of light they're been trained to associate as beneficial*. 

* Quickly skimming a later review it seems that this study was fraught with problems, however.

This is deeply frustratingly brief : the final four pages at the end of chapter three are what the whole chapter should have been about. Give me details of the experiments. Give me the criticisms and counter-claims. Don't waste me time with an enormous prelude instead ! Credit where credit is due, he does give references, so expect more on this in the future.

Having prematurely "established" to his own satisfaction that plants are probably intelligent, thinking beings, Calvo proceeds to tackle how this might be. This bit feels somewhat better. Obviously plants don't have nervous systems, and I'm entirely sympathetic to the idea that you don't need this for consciousness. But they do use electrical currents, with various external stimuli (light, gravity, touch and all the rest) being able to trigger measurable electrical impulses. And while they don't have nerves, they do have internal networks along which such signals might propagate.

There are two (!) other points about plants which are interesting. Apparently some parasitic vines behave in an extremely similar way to hunters. They have a sense of smell and can follow a trail, in time lapse sweeping back and forth just like an animal. They will adjust their hunting strategy based on what sort of trail they're following, showing a clear preference for tomatoes over wheat. 

The second point is that plants have a more developed sense of light than we might suppose. In a few cases they actually prefer darkness, apparently in response to the need to search for support structures (e.g. trees) on the forest floor. Weirdly, they can also sense colour (or at least wavelength), having preferences for different coloured poles over others.

Finally, Calvo points out that bacteria have been shown to respond differently as a result of their actions as they interact with the world, as though they're observing and learning. There's even evidence for Pavlovian conditioning. And I'll grant that if mere bacteria can do this, then it would be weird if plants couldn't, but I'd prefer to check the references before commenting any further.

The two most interesting things here to me are the possibly Pavlovian responses and the wavelength sensing. The Pavlovian behaviour I find interesting because it suggests something much more indirect than pure sensory stimulation, especially when it happens as a result of a completely unrelated activity to what the plant is seeking. This, if true, is at the very least some sort of learning; it might be a step too far to call it "reasoning", but it would be a step in that direction. The wavelength/colour issue might imply qualia. For that, the best I can come up with is that we'd need some way to create an illusion that would trick the plant into perceiving a colour that wasn't really there, but how we'd do this I don't know.

That's all there is, in the entire book, about behaviour that seems genuinely cognitive and not just automatic. It feels like scraps from the table at most.

2) The other stuff

While much of it is just standard introductory philosophy of mind stuff, there are some interesting bits in the rest of the book I do want to mention.

The first is a nice joke about vitalism :

"If you presented a dynamo to scientists of various disciplines, they would all look at it in different ways. A chemist might dissolve it in acid; a molecular biologists would take it apart and describe its components in detail. But, were you to suggest that 'an invisible fluid, electricity' flowed in the dynamo, the flow of which ceased when it was dismantled, they might 'scold you as a vitalist'."

Which is a reasonable enough point. Likewise, that biologists are apparently only just getting to grips with the idea that fish are sentient is incredibly worrying and bordering on the idiotic. Calvo is right that the flip side of anthropomorphism, which indeed we should avoid, is anthropocentrism : the idea that humans are uniquely special and superior. Both ideas would seem inherently and almost self-evidently wrong-headed; my news feed frequently comes with headlines about how scientists have "discovered" what all pet owners have known for years. I don't mean some nuance about animal behaviour, I mean bloody obvious things that they ought to be ashamed of for not realising decades ago. So I'm with Calvo on that score, especially as this seems thoroughly backed up my other recent reads.

But alas, that's just about where my sympathies end. It does not follow that, just because some ideas considered crazy in the past have become accepted into the mainstream today, that all crazy ideas deserve serious treatment. There's quite a strong whiff of the classic "they said I was maaaad !" vibe about the whole thing.

For instance, in yet another effort to encourage open-mindedness, he quotes a philosopher who bizarrely believes we can escape from Mary's Room. His analogy was... we can get used to the differences between the Celsius and Fahrenheit temperature systems. 

Da fuq ?

Riiiight. So I'll just go and imagine some new colours then, because apparently that's totally the same thing. Good grief.

And then Calvo doesn't do himself any favours by singing a paean to Martian robots but then saying it would be better if they could just grow from place to place for some reason. Good lord man, what are you talking about ? He goes off on one about the rights of plants but fails to explain what, exactly, if plants have awareness, are we ever supposed to eat without feeling massive, crippling levels of guilt, which would seem like the most obvious question in the world.

And finally there's a horrible anecdote that is unbecoming of professional scientists; here my sympathy returns partially to Calvo. As with the last book, there seem to be some appalling biases in the life sciences. For god's sake people, sort yourselves out.

"After out meeting, a group of leading plant physiologists published a paper attacking our work, without even the veneer of politeness usual in academic publishing. They argued that 'dubious ideas about plant consciousness can harm this scientific discipline', and 'generate mistaken ideas about the plant sciences in young, aspiring plant biologists.' We were not only wrong, they seemed to think, we were dangerous : 'serial speculationists' looking to dismantle respectable science from within. They urged funding agencies to refuse us and journals to reject our papers, to keep our 'prolific speculating and fantasising' out of scientific discourse."

Now some of this is inevitable : if you really think a discipline has been corrupted by pseudoscience, it needs to be called out. But you don't do it in personal terms in a paper. Just as I wish Calvo would shut up and make his actual claims, so the counterarguments should be restricted to just that. If you want to make more personal attacks, do it by all means*, but not in a journal. 

* Sometimes crackpots deserve to be treated as such.

Yet there are two sides to every story. I don't know about his papers but Calvo's book is absolutely 100% the work of a serial speculationist because there's so little bloody data in it. He might very well just be a nutter; the Michio Kaku of botany, perhaps. A very quick look at the paper in question and I think some of their claims are pointless and silly ("plant's just can't be conscious, mmkay ?") but some are very likely correct (evidence for specific experiments being wrong, evidence that Pavlovian responses can occur - surprisingly - without consciousness). From what I've skimmed, it may well be that the entire edifice of research into plant intelligence is without any foundation at all.

But for that, it seems prudent to reserve judgement and read the actual papers. Pending that, I might well revise my assessment of the book. As it stands, Calvo certainly hasn't done himself any favours with this at all.