— Man’s Search for Method
To succeed in the self-help business, you need three things: a good story, good-looking charts, and an acronym: “Here is how Stella solved her problem” “72% of all people have this problem” “Just follow the eight easy steps of the FURZWURM-method!”
For people to feel that they are getting valuable advice, you need something that feels like a mechanism. There is a feeling of grip and resolve, of closure and safety, once you have a sequence of steps to apply. Everything other feels inconclusive, vague, and unsatisfactory.
One irony is that it can sometimes be the mechanistic method itself that is unsatisfactory. For CVs, ads, or products — anything that competes — any best practices are actually self-defeating: if everybody followed their principles, doing so would precisely not stand out. The method becomes an entry condition, not a hallmark of success — lifting the threshold for everyone. Mechanisms are general. What a method does precisely not do, is make you understand your own specific case better. That’s its strength and its weakness: It allows you to act without thinking specifically about the situation at hand.
Mechanic methods pervade work, burocracy, and, courtesy of the self-help industry, ten thousands of how-tos for social and interpersonal situations. Sometimes method even gets confused with science. But of course there is some truth in the attraction of mechanistic thinking, and a world without it would be unthinkable. Still, method is also treacherous, and it betrays a pervading human weakness: the longing to abdicate judgement and responsibility. Method feeds our urge to stop thinking.
This happens especially as an institution matures. In a startup, everything is in flux, and everyone takes initiative. There is no clear division of labour, no processes, no policies, no precedents. But if the startup survives this, it starts building all of these. It has to.
Hiring is a good example. What thoughtful and observant discussions did we have when we made up our mind about a new hire. But then came the lists with ready-made boxes to tick and scores to calculate. It avoids bias, they say. True, bias is one of the problems of bad judgement. But there is a reason why we do not hire stricly by noselength or number of birthdays, though it would be perfectly unbiased: Relevance matters, too. And what is relevant does not stay stable from one human being to the next. General categories always introduce their own bias: towards the known and safe, against the new, the complex and the dialectical. Yes — open-minded judgements without rails are harder to make. So we better not make them, rather than learning to be better at it?
Methods are safe. Safe from being sued. Safe from being criticised for mistakes. Safe from making our own judgements or learning something new. By applying a method, you get the feeling that you are in control of the situation. But actually it is not you who is in control — the mechanism is. And sometimes that’s ok (it’s what is called reason). But also sometimes you are secretly happy to have relinquished control to the mechanism to become its advocate. And like with any advocate, the mechanism grants you power, while it blurs responsiblity. If it is right, you’re good. If it is wrong, it’s the fault of those who put it in place. But neither the mechanism, nor the individual who created it, are really responsible for the individual decision made here and now.
We were lucky so far: My department still hires without method. We still like to look at the individuals, even at the risk of making mistakes. You never know: Often you need new categories for every single applicant. What stands out? What would be the interplay between their talents and traits? How would they change us? Such discussions defy preconveived hiring criteria and the answers are not quantifiable. Matrices cannot replace human judgement. Our language has thousands of words for describing human beings, not just a dozen. But yes: using them all is not safe.
Ok, I know — I have called for robust procedures and structures for years. They do bring reliability and trust, efficiency and sanity. They also allow us to store experiences in them, so that not everyone has to reinvent the wheel all the time. But just like experience gets in the way of life in old people, policy gets in the way of a flexible organisation, and method gets in the way of being a thinking and responsive human being. Maybe this is the dialectics of methods: We cannot do without them. And we should never trust them, either.
There is a lot of philosophy we could touch from here: Kant’s distinction between understanding and reason, Hegel’s criticism of contemporary analytic philosophy appealing only to the former. And of course Teddy Adorno, in all his negative, dialectical beauty. He probably understood best that there cannot be the other, the genuinely unmethodical, the purely situative. The last thing I want is to foster that romanticism of the freewheeling and unvonventional, which some who like to feel like artists have made their identity. It often justifies lazi- and selfishness, or just making a stand for the sake of making it. As a universal claim, however, situativeness is absurd and empty. But we sometimes have to talk about method and thinking as if it were an antagonism.
So what I criticise here is not method. I criticise the longing and praise for it. The secret assumption that a method always knows better. That a method should be what we are striving for. That only a method provides a final answer, and a bulwark against bias and other errors.
It doestn’t.
What it does, however, is turn us into machines. We seem to be seeking to become mechanisms ourselves. We can only rest when we find a method, that frees us from needing to think. Maybe this attitude is especially fostered in current times: the times of coaches, counsellors and consultants — and computers. Because that’s how THEY work. In order for computers to work with us (AI or not), we need stable categories, quantifiable results, and to define clear steps. And given that software is eating the world, it would be strange if this would not also change our idea of what this world is or should be like. Calling a hotline, dealing with a clerk in a restaurant chain, or watching a few ads for fitness trackers should teach us that. When some time soon chat bots will reliably pass the Turing Test, it may well be that this is not just because the bot has become so human-like — but also because humans have become a lot more like a machine.
