Hindsight
The awesome power of hindsight.
|
JC: [Winding down a typically rambling rant] “... so, I ask you: why have we been so bad at stopping catastrophic risks from happening? Leeson, LTCM, Amaranth, Enron, Global Crossing, Kerviel, Madoff, Bear Stearns, Lehman, LIBOR, Theranos, London Whale, Mossack Fonseca, 1MDB, Wirecard ... ”
Middle manager: [interrupting]: Simply put, we were not proactively looking for it. We were not using data properly to evaluate risk. To add value as an in-house legal function we need to use innovative tools to crunch data, proactively spot emerging risks and escalate them to business.”
On every disclaimer any financial services firm has ever sent — and ever will send — however paranoid, overblown, absurd, weird or wonderful, there will be one constant, ringing reminder: one thing, which, above all the other
A cold night in Vienna, 1808
On a freezing night December 1808, about 1,500 people attended an akademie concert at the Theater an der Wien in suburban Vienna. The programme was to be four hours long, during which a young composer from out of town[1] would be debuting a number of works.
The run-up to the concert did not bode well: many of the musicians in theatre’s house orchestra had “conflicting commitments” with a better-paying gig across town at the Burgtheater, and even the solo soprano dropped out at the last minute, to be replaced by an unknown teenager— “I have to hop” is no modern excuse — and the composer, an irascible fellow, kept changing the music right up to the last minute. So poor were his relations with his musicians that, on some accounts, they refused his baton, and another conductor was drafted in to lead the orchestra on the day of the concert.
The concert was a disaster. What orchestra he could scrape together was under-rehearsed — “lacking in all respects”, according to one reviewer — the poor young soprano suffered stage fright, the hall was freezing and the show badly overran. During one work, the orchestra fell apart completely and had to restart the piece altogether.
So why, two hundred years later, is there even a record of this concert — most of Supercheese’s concerts were like that, and there’s no record of any of them — and how has it found its way onto a Jolly Contrarian article about hindsight? You will not be surprised to hear there is a punchline.
Of the scathing reviews that followed, one at least — in the Allgemeine musikalische Zeitung — was prescient enough to say the following: “To judge all these pieces after only one hearing, especially considering [...] that so many were performed in a row, and most are so grand and long, is impossible.”
To the punchline then: the young foreign composer was, of course, Ludwig Van Beethoven, and in that one concert he premiered his Symphony No. 6 (“Pastoral”), his Piano Concerto No. 4, and the aforementioned Choral Fantasia, as well as playing a few choice cuts from his Mass in C Major. If that wasn’t enough — and surely the premiere of the Pastoral, by itself, would have been enough to make the record of humankind’s highest achievements — after the interval, the orchestra debuted the most revolutionary piece of music, bar none ever written: the Disco Theme to Saturday Night Fever,[2] although then known only as Symphony No. 5 in C Minor.[3]
Okay, so, hindsight?
If the JC could travel back in time for one night only, throughout all of human history, this is the time and place he’d choose. He’d take thermals and a cushion. Imagine being one of those lucky 1,500 who heard the fifth symphony for the first time in history. There are few profound watershed in the cultural history of our civilisation, but that is surely one. But it would be imbued with such significance only because of the two hundred years of subsequent history from which we now benefit. Those lucky 1,500 probably found it quite tiresome. They may well have — most probably did — come in later life to freight that experience with more meaning than they could apprehend at the time.
This is the human condition, summarised. Only once it has happened — past tense — and often, only months or years after that, can we possibly apprehend the significance of unexpected events.
The category error: providing for the future by reference to the past
This is where my friend the middle manager makes his category error. Data all come from the same place: the past. When we review risks, catastrophes and step changes; when we consider punctuations to the equilibrium, fair or foul our wisdom, our careful analyses, our sage opinions, our hot takes, our thought leadership — all of these are derived from, predicated on, and delimited by data which, at the time the event played out, we did not have.
And herein lies the tension and profound dilemma of the received approach to modern legal practice. For we commercial lawyers are charged with anticipating the future and but sent out to battle armed with a methodology drawn exclusively from the past. Just occasionally, this dissonance rears up and hits us: the financial controller who must hold capital against shortcomings in a contract documenting a transaction that has already matured — a risk that has not come about, even though the reporting period rubles on: the formal imperatives of legibility taking priority over the logic of common sense.
It is a persistent frame: by the time we get around to analysing any catastrophe and how it played out, all circumstances are known, all options have crystallised and all discretions have hardened. The employee who, with imperfect information and in the fog of war, tacked to starboard when hindsight revealed a safe harbour to port finds himself short an very ugly option which those with executive responsibility for his performance will be mightily tempted to exercise. This is the lesson of Sidney Dekker’s The Field Guide to Human Error Investigations: we blame the meatware, because it exonerates the executive.
What caused the catastrophe — an inquiry one can only launch upstream — forms, informs, and reforms the downstream narrative. Take Madoff: now we have the book, the film, the exposes, the fabulously gruesome congressional committee hearings everything about ’s investment strategy now is obviously bogus. There can be only one explanation: someone was asleep at the switch — someone further down the chain. This is where the historian’s perspective is more useful than the thought leader’s. We know Madoff’s regulators were not asleep at the switch because we have Template:AuthorHarry Markopolos’s testimony that he kept prodding them in the ribs and pointing out exactly the anomalies they should have, the now agree in hindsight, been alarmed by.
it wasn't that the operators were asleep, that is to say, but that their switch didn’t work. But a maladroit switch is the responsibility of the executive, not the worker.
As of January 2021, we have an opportunity to see this dissonance live, in the wild, before it to ossifies into wise hindsight, with the GameStop phenomenon. Here, “lack of attention to data” is less plausible even than usual because the putative “victims” — hedge fund industry titans: tiny violins, right? — had arguably more data, and more data-processing capability at their disposal than anyone in history. Their opponents, a rag-tag aggregation of retail deplorables coalescing around a Reddit channel, had the network effect and the wisdom of crowds in their corner, but nothing like the computer horsepower nor institutional support of the funds they took down.
A “failure to crunch data” was not the problem here. The problem was a failure of narrative. Part of that failed narrative was the primacy of data. In God we trust, all others must bring data. But will that be the lesson we all learn from this debacle? Like the certain death of a bricks-and-mortar company with an irretrievably broken business model, don’t bet on it.
It is interesting to that when executives appeal for change, for “revolution” — ironic, I know, but I’ve seen it happen — for “a new way of working” they are not talking about the revealed failings of their own narrative — it's got them where they are, after all, so it can't be all bad — but is upon the inconstant performance of those mortal, expensive, fallible meat-sacks snoozing away at the switch. Thanks to overwhelming confirmation bias the idea that the switch isn’t working, and it’s the hierarchy supporting the broken switch that is not fit for purpose, somehow fails to occur.
See also
- Signal-to-noise ratio
- Proactive
- GameStop
- The SEC general counsel trying to plead executive immunity from testifying before Congress
References
- ↑ Bonn, in northwestern Germany — a long way from the cultural capital of the Austrian Empire.
- ↑ I am sorry. I couldn’t resist.
- ↑ Anyone interested in Beethoven’s symphonies — that is, in Jimi Hendrix’s words, “everybody here with hearts — any kind of hearts — and ears” — should check out Professor Robert Greenberg’s lectures about Beethoven.