The importance of experimentation and failure in improvement

Making madeleines was my practical experience of experimentation and failure in improvement

I write fairly often here about the importance of experimentation and failure in improvement, because I believe both are vital in refining your work. Today I’m doing it again, but I’ve got a personal example to share, because I think it’s important too that you see that I try to practise what I teach! I’m also sharing this example in detail because it gives you an idea of how Alexander Technique thinking looks ‘in the wild’.

The background to experimentation and failure in improvement

FM Alexander’s whole approach to organising thinking and movement had its roots in experimentation and failure. He spent months watching himself in a mirror (sometimes 3) as he recited. He observed, he made hypotheses, he tested them. The first chapter of his book The Use of the Self, entitled ‘Evolution of a Technique’ is a frequently detailed description of the way he experimented to relieve his vocal hoarseness:

… at least I could do no harm by making an experiment. [1]

I realised that here I had a definite fact which might explain many things, and I was encouraged to go on. [2]

I continued with the aid of mirrors to observe the use of myself more carefully than ever… [3]

I would give the new directions in front of the mirror for long periods together, for successive days and weeks and sometimes even months… [4]

Alexander also experienced a huge amount of failure in the midst of his experimentation, and periods when he gathered data that didn’t help to advance his thinking. And sometimes he did feel discouraged, but he didn’t allow this to impede his work.

I practised patiently, month after month, as I had been doing hitherto, with varying experiences of success and failure, but without much enlightenment. In time, however, I profited by these experiences… [5]

I firmly believe that if we are to truly learn from Alexander’s work, we must also take on board his example with regard to the role of experimentation and failure in improvement. Quite simply, you can’t improve without changing, and in order to change you have to allow for the possibility of failure.

The Great Madeleine Disaster of 2019

Last week I fancied making some madeleines. I have a nice tin that I bought in France, and I don’t use it as often as I’d like. I also had found a new recipe that I fancied trying – it didn’t follow the same procedure as my trusty normal recipe, and it added honey. It sounded like fun. Out came the tin and the ingredients.

I halved the recipe – I didn’t need masses of the things. And I had to bake in two batches, because the tin is small. The first batch was unsuccessful. The madeleines spread rather than rose, and they stuck to the tin. After digging them out. I paused and had the following thoughts.

Analysis 1: They stuck A LOT.
Hypothesis 1: I didn’t grease the tin sufficiently.
Test 1: Give the tin a really careful greasing, and a careful coating of flour to prevent sticking.

Analysis 2: They spread A LOT.
Hypothesis 2: This is because of the honey – it tends to cause that sort of spread pattern when added to baking. Alternatively, it might have been caused by the odd mixing method in the recipe. Hard to tell which at this point.
Test 2: throw in a little baking powder to see if that counteracts the spreading. If it’s the honey, it should give a sufficient lift to help. If it’s the odd method, it should make up for the lack of the introduction of lightness and air in the mixing.

So I tried both those things on the second batch.

Madeleines, Take 2

The second batch were even worse than the first. They still spread, but not as much. They rose up stunningly well, and then collapsed back down to create a crisp exterior and a raw interior. They were totally inedible. On the plus side, they didn’t stick to the tin! I had a good think, and these were the results of my analysis:

Analysis 1: Careful greasing of the tin was a big success. Go me!!

Analysis 2: The rising and falling pattern happens when there is either too much raising agent, or the oven is too hot.
New hypothesis: the oven temperature was too high.
Test: check against other recipe.

Sure enough, when I checked my usual recipe, the oven temperature was a lot lower. So I learned some really important things:

  • Grease the tin very carefully indeed
  • Make sure the oven temperature isn’t too high
  • The traditional mixing method for madeleines helps given them lift. If adding honey, use the traditional mixing method because it will help counteract the honey’s ‘spread effect’.

Experimentation and failure: vital tools

It’s never nice to have a baking failure. But this one taught me a lot about things I need to consider in order to make my baking better than it was before. And that’s the whole point about trying things and failing: from analysing the failure you learn things that you didn’t know before. You refine your knowledge of technique and principle. You learn to apply them more carefully. And when you do these things, you become better at what you do. So don’t be afraid of experimentation, and enjoy your failures. Your baking will be better for it.

[1] Alexander, F.M. (1985[1932]) The Use of the Self London: Orion, p.26.

[2] ibid., p.28.

[3] ibid., p.33.

[4] ibid., p.41.

[5] ibid., p.32.

Image by Varaine [CC BY-SA 4.0 (]