I am often asked why I do not recommend soy protein for vegetarians.
On paper, using the logic of “nutrientism” (the idea that we should focus only on isolated nutrients in a food, and that all “nutrients”, e.g. “protein” or “vitamin C”, are equivalent and interchangeable), soy makes sense. It is a plant that appears to be high in the types of amino acids that humans require. It’s relatively easy to grow and is sort of like the universal solvent — you can make damn near anything from it.
In real life, it’s not so simple. “Nutrients” are not whole, complete, discrete foods. Our bodies respond to whole foods, eaten as part of a whole diet, in a context of a whole life and whole environment. We eat food. Not “nutrients”. And, as a food, soy doesn’t offer much benefit compared to its drawbacks.
The aggregate data suggest that whatever benefit soy may offer is vastly outweighed by its many liabilities — especially when it is processed, as it must be in order to be edible. (Really guys — did you honestly think that Tofurkey was healthy?) For a review of the clinical literature and concerns, check out The Whole Soy Story.
One key problem with soy appears to be its effects on brain health with aging.
A study in the JACN followed thousands of subjects following a variety of Japanese and Western-style diets in Hawaii. The study concluded that in subjects ranging from their 70s to their 90s, “poor cognitive test performance, enlargement of ventricles and low brain weight were each significantly and independently associated with higher midlife tofu consumption.” In other words, the more tofu a person consumed earlier in life, the worse their prognosis for healthy brain aging would be.
Of course, this isn’t causative, merely correlative. Other studies have suggested that soy isoflavones may offer a protective effect. Again, the data is mixed.
For instance, one study points out that our conclusions about healthy brain aging depend on the indicators we use to measure it, as well as sex differences. Given soy’s estrogenic actions, what effects does it have on men and women? (In women, for instance, should we assume that more estrogenic activity is always good?)
Another study points out that we have to distinguish between soy’s effects in vitro (i.e. on cell cultures in a petri dish) and in vivo (i.e. in actual people living real lives). “Lab significant” may not be “real world significant”, and/or effects on free-living people may be much different than effects on cells bathed in a medium of isolated substances. As this study comments, “While it has been shown that the soy phytoestrogen genistein inhibits neuroprotective functions in cell cultures, recent in-vivo findings strengthen the case for a possible causal mechanism of soy-induced neurodegeneration.”
Now, the JACN study is ten years old, but another study from 2008 offers similar but somewhat more nuanced results. High tofu consumption was associated with worse memory, while high tempeh consumption (a fermented whole soybean product) was independently related to better memory, particularly in participants over 68 years of age. Another study looking at the same population found that the effects changed when the groups were disaggregated by age and type of memory recall. (BTW, both studies found that fruit consumption was also associated with improved memory across the board.)
What many of these studies indicate to me is that “soy” is not a homogeneous category, just like “meat” can encompass everything from rare Kobe beef to baloney.
Having an ounce or two of traditionally fermented tempeh now and again — importantly, within the context of a traditional East Asian diet, eating patterns, and lifestyle — is not the same thing as drowning in TVP, soyloney, soy milk, and soy cereal every day within the context of a modern Western diet, eating patterns, and lifestyle.
Ask yourself: How is soy protein powder produced? How do they make soy bacon? Have you read the labels? After considering the actual process by which most mass-produced soy products are created, do you still want to eat them? Do you still consider them “healthy”? What does the rest of your diet look like?
Thus, when we ask “Is soy bad/good for me?” we have to ask:
“Which soy? In what format?”
“For what purpose?”
“How much are you consuming?”
“What exactly are you assuming that soy will do/not do for you?”
“In whose interests would my soy consumption be?” (In other words, who is telling you soy is good and to eat more?)
“WTF is Tofurkey and why is this offense to Saint Julia Child on my table!?”