Did Edison have a peer reviewed study supporting the techniques he was using to create the first lightbulb?
Similarly, did the first person to ever bake a cookie already have a recipe for it?
Imagine being the person that did bake the first cookie. It would be amazing – by some stroke of luck, you’ve happened upon something entirely new and delicious! Imagine being proud of this accomplishment and taking the cookie to one of your friends for them to try it.
They take a bite and say, “Wow, that’s interesting, I haven’t had anything like this before. That tastes really good. What’s the recipe?”
And you respond, “There’s no recipe, I was just trying stuff out.”
“Oh, there’s no recipe?” They frown slightly.
“No, there’s no recipe… I was just trying things out, and this was one of the results,” you say, a bit confused. “I’ll keep trying to make it better and then eventually there might be a recipe, but this is where it is right now, and I think it’s very good so far.”
“Hm. I’m sorry, I’ll admit it – this thing tastes pretty good, but I only eat things that have recipes,” they say matter-of-factly, crossing their arms. “No matter how good it tastes, if it doesn’t have a recipe, it’s not good enough to eat.”
You walk away, completely baffled. How weird was that? Do you think that exchange ever would have happened? Of course not. The insistence of your friend on having a recipe (evidence) in order to validate your results – and consider them legitimate – makes absolutely no sense. If you were to write a recipe, that could only ever occur after you’ve successfully baked the dang thing. The first ever cookie was certainly not the result of rigorous scientific evidence. It was probably a somewhat serendipitous outcome that came about from someone who was playing around with an assortment of possibilities. Thus, the discovery of the cookie (among many other things) and the subsequent recipe-writing that presumably took place were most likely the result of tinkering.
Author, philosopher, and investor Nassim Nicholas Taleb coined the term “Stochastic Tinkering,” and it’s one of the most powerful concepts I’ve come across when thinking about how innovation and new ideas are formed, especially in the context of our current obsession with being evidence-based. (Note: I am not anti-scientific evidence. I am simply anti-false dichotomy. More on this later.)
Put most simply, tinkering can be boiled down to discovering the “best fit” solution or idea as the product of two elements: (1) some degree of randomness over time and (2) perceiving whatever new or surprising changes occur as a result of this randomness.
On a personal scale, this amounts to running small experiments over a period of time – maybe manipulating a few factors here and there – and observing what ends up being beneficial in dealing with the randomness of the world. We can then use these new or surprising changes to continue tinkering and make further observations. This process can be repeated until a desired conclusion is reached, but the caveat here is that you must be comfortable not knowing what kind of conclusion might be reached. You may have an endpoint in mind, but a true tinkerer knows that it may manifest in a totally different way than what they had envisioned.
Now, what’s interesting from my own observations is that it’s almost exclusively the case that the established evidence (e.g. studies) follows the tinkering – not the other way around. The former ends up substantiating the latter, not somehow canceling it out, as some seem to think it does.
The history of bodybuilding, starting with Eugen Sandow (often credited as the father of modern bodybuilding) is another perfect example of successful tinkering leading to the establishment of evidence. To answer the question “What is the best way to build muscle?” in a time before scientific rigor, people just tried different things, sometimes by accident. There’s the element of randomness. Then, when someone happened upon something that worked – a new or surprising development discovered by Sandow, evidenced by his strength and musculature – people flocked to it, and wanted to learn his techniques. And plenty of people were able to gain muscle and strength by doing so.
Nowadays, because of the impetus issuing from continuously established anecdotal evidence (i.e. we became interested in studying it precisely because people were making gains in strength and muscle for decades with no studies to back anything up), we have a plethora of mechanistic studies, controlled trials, and meta-analyses delineating which training, nutrition, and recovery techniques are best for gaining muscle and strength. There is nothing wrong with this explosion of scientific data, per se. Again, I only implore you to observe the order in which these events took place, and the implications of this order for the current medical/academic obsession with being purely “evidence-based.”
Did the lack of research in Sandow’s time invalidate all the results gained before the research existed? Of course not. That wouldn’t make any sense, in the same way that not having a recipe for the first ever cookie didn’t invalidate its flavor. However, now that so many of us have given scientific studies unequivocal primacy, people believe anything unsubstantiated by current research to be inherently untenable. In having this belief, we run the risk of throwing the baby out with the bathwater. Think about this: if it weren’t for forward-thinking people tinkering with new ideas and concepts outside of the established evidence, then we would never have any innovation at all.
It may seem that this concept introduces a science vs. tinkering dichotomy. However, I can assure you this is a false dichotomy. In a way, tinkering is scientific, as it can be used for a sort of fluid hypothesis testing, but it just doesn’t require controlled trials for you to observe the effects. This lowers the scientific rigor in a sense, but individuals need to ask themselves just how important scientific rigor is when applying things to their own lives. You don’t really need a controlled trial to decide that you feel better, look better, perform better, etc. You just need your own experience. And just like research doesn’t invalidate experience, your experience doesn’t invalidate research, either. Although many people treat them as if they are at odds, they are not – nor do they have to be.
If there is tension between what has been established in the literature and your personal results, then there is something to be explored and learned there.