What the inven­tion of com­plic­ated foods tells us about dis­cov­ery, innov­a­tion, and uni­ver­sity research funding.

Over lunch earlier this week we got talk­ing about how dif­fer­ent foods get dis­covered — or inven­ted, whichever’s the most appro­pri­ate model. The point of the dis­cus­sion was how unlikely a lot of foods are to have actu­ally been cre­ated in the first place.

The lin­eage of some quite com­plic­ated foods is fairly easy to dis­cern, of course. Bread: leave out some wet flour overnight and watch it rise to form sour­dough. Do the same for malt and you get beer (actu­ally the kind that of beer that in Flanders is called lambic). Put milk into a bar­rel, load it onto the back of a don­key and trans­port it to the next town, and you’ll have naturally-churned but­ter. It’s fairly easy to see how someone with an interest in food would refine the tech­nique and diver­sify it, once they knew that the basic oper­a­tion worked in some way and to some degree.

But for other foods, it’s exactly this ini­tial step that’s so problematic.

I think the best example is meringue. Con­sider the steps you need to go through to dis­cover that meringues exist. First, you have to sep­ar­ate an egg — which is obvi­ous now, but not so obvi­ous if you don’t know that there’s a point to it. Then you need to beat the white for a long time, in just the right way to intro­duce air into it. If you get this wrong, or don’t do it for long enough, or do it too enthu­si­ast­ic­ally (or not enthu­si­ast­ic­ally enough) you just get slightly whiter egg white: it’s only if you do it prop­erly that you get the phase change you need. Of course you’re prob­ably doing this with a wholly inap­pro­pri­ate instru­ment — like a spoon — rather than a fork or a bal­loon whisk (which you don’t have, because nobody knows there are things that need air beat­ing into them yet). Then you need to determ­ine, counter-intuitively, that mak­ing the egg white heav­ier (with sugar) will improve the final res­ult when cooked. Then you have to work out that cook­ing this liquid — which has actu­ally to be a pro­cess of dry­ing, not cook­ing — is actu­ally quite a good idea des­pite appearances.

It’s hard enough to make a decent meringue now we know they exist: I find it hard to ima­gine how one would do it if one didn’t even know they exis­ted, and fur­ther­more didn’t know that beat­ing egg whites in a par­tic­u­lar way will gen­er­ate the phase change from liquid to foam. (Or even know that there are things called “phase changes” at all for that matter.)

Think­ing a little harder, I actu­ally can ima­gine how meringues got inven­ted. In the Middle Ages a lot of very rich aris­to­crats com­peted with their peers either by knock­ing each other off horses at a joust or by exhib­it­ing ever-more-complex dishes at feasts. These dishes — called sub­tleties — were inten­ded to demon­strate the artistry of the chef and hence the wealth and taste of his pat­ron, the aris­to­crat. Pies filled with birds, exact scale mod­els of castles, work­ing water-wheels made out of pastry, that kind of thing. In order to do this sort of thing you need both a high degree of cook­ing skill and a lot of unusual food-based mater­i­als to work in. You can find these as part of your nor­mal cook­ing, but it’s prob­ably also worth some exper­i­ment­a­tion to find new and unusual effects that will advance this cal­or­ific arms race a little in your favour.

So maybe meringue was inven­ted by some medi­eval cook just doing ran­dom things with food­stuffs to see what hap­pens. The time spent on things that don’t work — leav­ing pork fat out­side to see if it fer­ments into vodka, per­haps? — will be amort­ised out by the dis­cov­ery of some­thing that’s really use­ful in mak­ing really state-of-the-art food. Con­trary to pop­u­lar belief the Middle Ages was a time of enorm­ous tech­no­lo­gical advance, and it’s easy to think of this hap­pen­ing in food too.

So food evolves under the com­bined effects of ran­dom chance oper­a­tions shaped by sur­vival pres­sures. Which is exactly what hap­pens in bio­logy. A new com­bin­a­tion gets tried by chance, without any anti­cip­a­tion of any par­tic­u­lar res­ult, and the com­bin­a­tions that hap­pen to lead to decent out­comes get main­tained. At that point the bio­lo­gical ana­logy breaks down some­what, because the decent out­comes are then sub­jec­ted to tele­olo­gical refine­ment by intel­li­gent beings — cooks — with a goal in mind. It’s no longer ran­dom. But the ini­tial undir­ec­ted explor­a­tion is abso­lutely essen­tial to the pro­cess of discovery.

Bizar­rely enough, this tells us some­thing more gen­eral about the pro­cesses of dis­cov­ery and innov­a­tion. They can’t be goal-directed: or, more pre­cisely, they can’t be goal-directed until we’ve estab­lished that there’s a nug­get of prom­ise in a par­tic­u­lar tech­nique, and that ini­tial dis­cov­ery will only be per­formed because of someone’s curi­os­ity and desire to solve a lar­ger prob­lem. “Blue-skies” research is the start­ing point, and you by defin­i­tion can’t know — or ever expect to know — what bene­fits it might con­fer. You have to kiss an awful lot of frogs to have a reas­on­able expect­a­tion of find­ing a prince, and blue-skies, curiosity-driven research is the pro­cess of identi­fy­ing these proto-princes amongst the horde of equally unat­tract­ive altern­at­ives. But someone’s got to do it.