[Update 8/2011: as I've learned more about human genetics and evolution, I've come to appreciate that many Europeans actually descend from early adopters of agriculture more than they descend from the hunter-gatherers that previously occupied Europe. Also, 10,000 years has been long enough for significant genetic adaptation. Read The 10,000 Year Explosion for more information].
You've heard me say that I believe grains aren't an ideal food for humans. Part of the reason rests on the assertion that we have not been eating grains for long enough to have adapted to them. In this post, I'll go over what I know about the human diet before and after agriculture, and the timeline of our shift to a grain-based diet. I'm not an archaeologist so I won't claim that all these numbers are exact, but I think they are close enough to make my point.
As hunter-gatherers, we ate some combination of the following: land mammals (including organs, fat and marrow), cooked tubers, seafood (fish, mammals, shellfish, seaweed), eggs, nuts, fruit, honey, "vegetables" (stems, leaves, etc.), mushrooms, assorted land animals, birds and insects. The proportion of each food varied widely between groups and even seasons. This is pretty much what we've been living on since we evolved as a species, and even before, for a total of 1.5 million years or so (this number is controversial but is supported by multiple lines of evidence). There are minor exceptions, including the use of wild grains in a few areas, but for the most part, that's it.
The first evidence of a calorically important domesticated crop I'm aware of was about 11,500 years ago in the fertile crescent. They were cultivating an early ancestor of wheat called emmer. Other grains popped up independently in what is now China (rice; ~10,000 years ago), and central America (corn; ~9,000 years ago). That's why people say humans have been eating grains for about 10,000 years.
The story is more complicated than the dates suggest, however. Although wheat had its origin 11,500 years ago, it didn't become widespread in Western Europe for another 4,500 years. So if you're of European descent, your ancestors have been eating grains for roughly 7,000 years. Corn was domesticated 9,000 years ago, but according to the carbon ratios of human teeth, it didn't become a major source of calories until about 1,200 years ago! Many American groups did not adopt a grain-based diet until 100-300 years ago, and in a few cases they still have not. If you are of African descent, your ancestors have been eating grains for 9,000 to 0 years, depending on your heritage. The change to grains was accompanied by a marked decrease in dental health that shows up clearly in the archaeological record.
Practically every plant food contains some kind of toxin, but grains produce a number of nasty ones that humans are not well adapted to. Grains contain a large amount of phytic acid for example, which strongly inhibits the absorption of a number of important minerals. Tubers, which were our main carbohydrate source for about 1.5 million years before agriculture, contain less of it. This may have been a major reason why stature decreased when humans adopted grain-based agriculture. There are a number of toxins that occur in grains but not in tubers, such as certain heat-resistant lectins.
Non-industrial cultures often treated their seeds, including grains, differently than we do today. They used soaking, sprouting and long fermentation to decrease the amount of toxins found in grains, making them more nutritious and digestible. Most grain staples are not treated in this way today, and so we bear the brunt of their toxins even more than our ancestors did.
From an evolutionary standpoint, even 11,500 years is the blink of an eye. Add to that the fact that many people descend from groups that have been eating grains for far less time than that, and you begin to see the problem. There is no doubt that we have begun adapting genetically to grains. All you have to do to understand this is look back at the archaeological record, to see the severe selective pressure (read: disease) that grains placed on its early adopters. But the question is, have we had time to adapt sufficiently to make it a healthy food? I would argue the answer is no.
There are a few genetic adaptations I'm aware of that might pertain to grains: the duplication of the salivary amylase gene, and polymorphisms in the angiotensin-converting enzyme (ACE) and apolipoprotein B genes. Some groups duplicated a gene that secretes the enzyme amylase into the saliva, increasing its production. Amylase breaks down starch, indicating a possible increase in its consumption. The problem is that we were getting starch from tubers before we got it from grains, so it doesn't really argue for either side in my opinion. The ACE and apolipoprotein B genes may be more pertinent, because they relate to blood pressure and LDL cholesterol. Blood pressure and blood cholesterol are both factors that respond well to low-carbohydrate (and thus low-grain) diets, suggesting that the polymorphisms may be a protective adaptation against the cardiovascular effects of grains.
The fact that up to 1% of people of European descent may have full-blown celiac disease attests to the fact that 7,000 years have not been enough time to fully adapt to wheat on a population level. Add to that the fact that nearly half of genetic Europeans carry genes that are associated with celiac, and you can see that we haven't been weeded out thoroughly enough to tolerate wheat, the oldest grain!
Based on my reading, discussions and observations, I believe that rice is the least problematic grain, wheat is the worst, and everything else is somewhere in between. If you want to eat grains, it's best to soak, sprout or ferment them. This activates enzymes that break down most of the toxins. You can soak rice, barley and other grains overnight before cooking them. Sourdough bread is better than normal white bread. Unfermented, unsprouted whole wheat bread may actually be the worst of all.