I have a thesis. Given any food you might care to name, someone has written, with certainty and rectitude, that it would be better to consume strychnine than to eat one portion of the nominally-edible substance in question.
I was not surprised to learn that the dining public often disagrees with nutritionists on what foods are healthy. Or, for that matter, that nutritionists themselves cannot agree on the same question.
But when we ask the question “Is (pick your food) healthy?”, what question is being asked and answered? We can be pretty sure it isn’t the question as stated, because that really isn’t a question at all – it’s a pseudo-question: ambiguous, biased, and constrained.
No one means to ask an imperfect question. That’s one of the problems with dicey questions – they don’t wear signs. This question, however, is not good – let’s imagine for a minute that I ask my computer “Is sushi healthy?”, and look at some of the problems.
First, not all sushi is created equal. Sushi – the core ingredient being vinegar-flavored rice – may have fish, or consist purely of vegetable ingredients. So right away my question is ambiguous.
Second, there is a plentiful serving of economic bias and constraint in my question: some people – perhaps most, if we cast a worldwide net – don’t have access to all the foods that would be best for them, or simply cannot afford them. Here in the US, all it takes is a trip to the local food bank for conformation of that. Even in the 21st century, we must eat what we can find, and we have to eat something.
Third, when I ask my computer “Is sushi healthy?” human diversity plays a role. I want an answer for me, not someone with a different health situation. However my browser doesn’t know me, so I am going to receive answers by and for other people. It may be difficult or impossible to know which answers apply to me.
Finally, people will quietly interpret “Is sushi healthy?” as other questions, ranging from their own preference “Do I like this food?”to “What is the worst thing that might happen if I ate this stuff regularly?” None of these is immediately helpful, even if I know which question is being addressed.
It’s not easy to sort all of this out, but that has now become my job – and I’m just a person asking a question – in systems parlance an “end user.” And if I knew enough to unravel this, I probably wouldn’t have asked the question in the first place.
The result? I may punt, and go with the answer I hoped for in the first place- that is, apply confirmation bias. But then I really didn’t answer any question – either the stated one “Is sushi healthy”, which has many answers, or a question close to my true interest, such as “How much sushi should I personally eat?” I just picked an answer out of the pile that I already liked, and fooled myself into thinking I answered the question I cared about.
What’s the right question? If you ask me, it is the same as it would have been 50 years ago: we see a professional we trust, and get recommendations. Self-help sites can be useful as background, but for most of us that’s about it – when we don’t know the questions being answered, we can’t trust the answers we receive. Our job in a consultation is to be sure we do identify and understand the questions being addressed, and that our personal constraints are taken into account.
It’s very common to ask one question, have data addressing other question, and to trick ourselves into thinking we answered the first question after all. As professional analysts, or just people trying to learn what would be healthful for tonight’s dinner.