He seems to be very conscious that this is in effect an experiment, that might be risky as well. E.g. from his site ( http://robrhinehart.com/):

"I am reticent to provide exact brand names and instructions because I am not fully convinced of the diet's safety for a physiology different than mine. What if I missed something that's essential for someone of a different race or age group"

But he is interested in actually gathering more data, even if not exactly formally enough for a proper study:

"So…I'll just ship you some of my batch. If you are willing to consume exclusively soylent, and get a CBC, chem panel, and lipid blood test before and after the week and share your results with me it's on the house. Bonus points for getting a psych evaluation before and after. The brain is an organ. I can ship it worldwide but it would be nice if you were in San Francisco so we can meet in person."

Further down on his blog, he has also posted PDF's of his bloodwork.

While he certainly does not have sufficient evidence of general safety and effect, he is aware of that and seems to be actively interested in learning about any flaws in what he's done so far.

> The burdon of evidence should be even greater when it's a non-natural source, as Nassim Taleb argues in Antifragile

This, to me, is ridiculous. A vast range of dangerous poisons are readily avaiable in nature.

I used to pick a lot of mushrooms, and can safely identify perhaps a dozen types common where I grew up. But every year there are people who die agonizing deaths from consuming various nasty toxins because they didn't pay close enough attention to what they picked. And the safety of most mushrooms is unknown - we simply don't have data, and the effects can take a long time to show.

In some cases, popular mushrooms are known to be toxic, such as false morels that contains gyromitrin. They are seemingly safe after boiling. Except there is still gyromitrin in the mushroom, just in small enough quantities that you're ok as long as you don't ingest too much. It's typical to recommend no more than one meal per season, as the poison is stored. But many mushrooms contains compounds we don't know the effect of, and where it is perfectly possible that no effect would be noticeable for a very long time - properly cooked false morels for example, might have no effect on you for years, until you get a bit careless with the cooking (a common way for people to get poisoned by false morels is to stand over the pot while boiling it...)

Nature has an abundant supply of horribly nasty toxic substances that might pass for food for a while.

This is before getting into what "natural" even means.

Everything has to prove its harmlessness. It's just that some things we have a lot of existing data for that at least demonstrates a certain level of safety.

But in the absence of data, I'm no more going to be willing to chew on some random "natural" substance than I'd be willing to ingest some random synthetic substance.

I really must defer to Taleb on this, I'll simply say he means natural in a more specific way than "all substances that exist in nature," and something more like, "the culture and patterns mankind has successfully practiced in relationship to nature for thousands of years with proven success."

That doesn't sound specific at all to me, but rather extremely vague and open to whatever interpretations might suit... What "culture and patterns" have we "successfully practised" "in relationship to nature for thousands of years"?

Diets vary crazily much, even within small geographical areas. Things like lactose intolerance is something we only "recently" started seeing on the wane, and it is wildly dependent on your heritage. I'm Scandiavian, and growing up I didn't even know there were people who had problems with dairy, as it was a total non-issue. I used to drink about a litre a day of milk on average growing up.... It was first as an adult I realised there are large parts of the world where people pretty much don't drink milk.

At the same time, I also consumed vast quantities of Scandinavian salted licquorice candy. Except most of if is ammonium chloride based. I doubt that would fit very well into the idea of something "natural", but it is an integral part of Scandinavian culture.

As is fish dissolved in caustic soda an integral part of Norwegian culture, and something we have "successfully practiced" for a very long time.

And processing such as hanging/drying, burying (until half rotted in some cases) and salting various types of food have been an essential part of human culture for a very long time.

And, yes, picking mushrooms and facing the risk of poison have been integral to our culture for a very long time as well, even though many of the most poisonous mushrooms keep claiming deaths up to this day.

Many of the traditional "natural" foods I grew up with are significantly less healthy than a modern heavily processed microwave meal, or the protein powder I use to supplement for my weight lifting.

This might be a different tack, but what if you take Mr. Taleb's "nature" to be human nature [1]? People don't tend to eat poison mushrooms as a rule.

In that case, his rule would be a good one. If you drift in to a new area of human endevour, the people who follow you out there are taking a risk until the area is well-traveled

[1] In the Aristotelian sense: human nature is the proportionate sum of all human activity. We mostly eat, drink, sleep, walk, talk, etc. so that makes up the bulk of our nature. Cannibalism and incest are acted out much, much less often, so that is a much, much smaller part of our nature, though it is tucked in there.

You define nature as activities that people perform most? By that reasoning the culture with the largest population is representative of typical human nature. That just seems wrong.

There are almost 2 billion Chinese. They love to eat chicken paws and cow intestines. So that is human's natural diet?

Chicken feet and cow intestines taste great. I'm not Chinese. My favorite dish when visiting Beijing a few years back was spicy duck intestines. Looked like noodles.

Yep, I'm saying if you generalize about all of us, that's how you'd do it.

Your point is dead on: if you make a generalization that broad, you lose a lot of information.

Not sure what makes chicken breast better than paws and cow rump better than intestines. All are animal parts.

But he doesn't have a 'miracle drink'. He has a generic food mash and his only claim is 'contains basic nutrients and calories'.

He says it contains all the nutrients the body needs. That is a big claim.

Given that he is asking for feedback from others who decide to try it, and on his blog warns that he is not sure how it might work for others, and is offering to provide samples to people who agree to have blood work done before and after trying it for a week, he seems acutely aware that while he thinks it contains all he needs, he does not yet have sufficient evidence to be sure.

He's not written a scientific paper - don't assume that every single word he has written is meant to be interpreted in the strictest way possible.

By all means be sceptical, but while he might very well be wrong, at least he appears to be earnestly looking to identify and correct any flaws which already puts him far above a lot of charlatans selling risky diets.

No, he says it has all of the ones we know about. He emphasizes the fact that there stands to be a lot of learning to do and bugs to work out.

It's a curious statement about our society where the idea that the fuel for the human body could be simple. "Basic nutrients and calories"

It's all produced by self replicating lifeforms we've only begun to comprehend, right?

Right, but those sophisticated replicators are so sophisticated that they tend to be rather tolerant of their input.

There's no reason to think we can't science out what that input needs to be, or at least to be able to find one or more valid subsets of input.

This is a rather large improvement over what most people do; one typically applies little to no science whatsoever. :-P

The liver is pretty tolerate of alcohol, but a lifetime of alcohol use can cause liver issues.

Right; the downside of a tolerant system is that it does not fail hard. This does not, however, preclude finding the system limits by use of the scientific method. It just makes it a tad trickier at times.

I'm getting the idea that people profess a "belief in science" in the general case, but then absolutely deny its applicability in the specific case here. ;-)

What is not natural with cigarettes?