Oliver Stadtler, in his book Japanese Pilgrimage, tells the story of the monk Kukai, the founder in Japan of the austere Shingon school of Buddhism. After his death, itinerant preachers turned him into the popular miracle-worker Kobo Daishi. Stadtler recounts how a number of local legends, originally told of other holy men, became attached to Kobo Daishi, a process which still continues.
In a similar way, one of the most famous quotes about mathematicians,
A mathematician is a machine for turning coffee into theorems
which was probably due to the Hungarian mathematician Alfréd Rényi, seems to have become attached to Rényi’s compatriot, collaborator and friend Paul Erdős. Many mathematicians and biographers seem to think that it was Erdős who said it. But an iconic quote like this is always likely to have multiple attributions.
A third Hungarian, Paul Turán, is said to have added the corollary, “Weak coffee is suitable only for lemmas”.
In any event, the popularity of the quote indicates that it does resonate with mathematicians, for reasons going beyond simple physiology.
In happy and productive mathematics departments, there is usually a ritual of gathering in a common space mid-morning and/or mid-afternoon. General conversation serves a social function; mathematical discussion spreads ideas and encourages collaboration; and grumbles about aspects of academic life reinforce collegiality and also let department heads and administrators know what we are thinking.
This is on my mind since, controversially, my department has decided to end the part-time employment of a person to make tea and coffee, replacing her with a machine.
The main argument is that this saves money, perhaps half a percent of our overall budget. Part of the controversy comes from the fact that the department has grown in size over the last decade, and there is a law which states that, as an organisation grows, the proportion of administrators also grows; administrative jobs grow too large for hard-pressed academics. Then inevitably people grumble when we can afford administrators but not a part-time coffee maker.
Our theorems do bring in money; we get research grants, and our citations feed into bibliometric funding formulae. That being so, if there is any truth in Rényi’s claim, then it makes sense to provide the highest possible grade of fuel. Machines can make decent coffee but woeful tea. (The reason is that, when I make a cup of tea, even at work, I boil a kettle and use boiling water; but the makers of vending machines don’t trust me to be near boiling water, presumably because they are afraid of being sued.)
Like the Nutri-Matic Drinks Synthesizer on the Heart of Gold in Hitch-Hiker’s Guide to the Galaxy, it produces a “cup filled with a liquid which [is] almost, but not quite, entirely unlike tea”. Fortunately, unlike that machine, it does not say “Share and Enjoy” when it dispenses a cup. But it tries to instil patience in its customers by waiting for a decent interval before condescending to provide change.
Then we come to biscuits. Bad for our health, but very comforting (especially when coated with chocolate), and important as social glue. And, if biscuits, why not doughnuts, as in some North American departments? Doughnuts can no doubt be turned into theorems by topologists.
The real problem is that the effects of coffee (better theorems, more collaboration, more collegiality) are not immediately obvious to administrators, and are not easily quantified (unlike the costs). But they are worth fighting for!