Wednesday, January 19, 2011

The power of Bullshit

The Department of Fisheries and Oceans has invented an esoteric language which it uses to assert its power. Photo

On January 17, the Cohen Commission resumed its hearings on the collapse of the 2009 Fraser sockeye. I decided to attend, because two DFO managers were called to witness on that day. Having heard the intense testimonies given by representatives of the First Nations back in December, I was curious. For three long days, aboriginal leaders had blasted the Department of Fisheries and Oceans in the most compelling manner. How would that agency respond? Would it send some of its top gun communicators in an attempt to control the damage?

It did not. Instead, it sent two archetypal technocrats who gave one the dullest and most abstruse testimonies on public record. It was dull to the point of being disturbing. Activists who have been following the Cohen proceedings have got into the habit of killing some DFO testimony time by playing “Bullshit Bingo”. The rules of the game are simple: you come with a card that has a grid of pre-filled bureaucratic words such as 'benchmarks', 'baseline monitoring', 'adaptive management', etc., you cross the words out whenever a DFO witness uses them, and when you cross out an entire line – Bingo!

Well folks, on January 17 DFO took the game of Bullshit Bingo to a whole new level. For the first hour, I was shell shocked. DFO Salmon Regional Resource Manager Jeff Grout was the first to speak. “One of the key elements that the forecasting model uses is a forward simulation which looks at historical spawning and recruitment data to try and understand what the performance might be of different harvest rules into the future”. “We use cumulative probability distribution”. “The relationships give a probability distribution on the range of possible returns”. “The probability of achieving the run is actually an inverse of what is being shown in this table”, etc., etc. for two straight hours. That man was a machine.

The mid-morning recess was welcomed as holy water by everyone, as it made the technobabble stop for a full fifteen minutes. This allowed my brain to reboot and start making sense of what was truly happening on that day. No one in this room understands what is being said, my brain whispered to my ear. Even the Commission's legal counsel who had taken on the tedious task of leading the questioning (and did a rather good job at that) appeared lost on occasion. The other lawyers were deep into their morning nap. As for the good Commissioner Bruce Cohen, he was frantically shifting his attention from the witness to the counsel to his notes and back to the witness again, like a trapped mouse in a lab cage which hits the same wall again and again in a desperate search for the exit.

And yet, those DFO bureaucrats sitting in the witness box were among those calling the shots in the field. They were making concrete life and death decisions over the salmon runs and the human and non-human communities depending on them. Those decisions, we learned that morning, were in large part based on “Frizzie”, a statistical forecasting model whose inner workings Mr. Grout was explaining to us. No doubt, this was a dry and arduous subject to grasp. But shouldn’t other people – shouldn’t I – make the effort of understanding it, given the enormous stakes involved? Isn’t the sharing of knowledge one of basic principles of our democracy? Bravely, I re-entered the court room after the recess, resolute on understanding what I was hearing.

But soon, my brain was stunned to complete numbness again by the sound of Mr. Grout’s voice. At one point the Commissioner’s counsel asked almost casually, as if to give us all a little breather, why DFO’s forecasting model used a 48-year time frame in the future. His answer:

In terms of the modeling work, there is a number of uncertainties associated with the model, including uncertainties about what the best model parameters would be to describe the population dynamics of these populations. There may be patterns in the annual abundance of the spawners that may change in the future, associated with a particular harvest rule. So we were wanting to look at the performance over a longer time frame to see what we might expect to have occurred. Through the various workshops, I think we also looked at different time frames during the planning period as well, but – One of the other reasons is that using a longer time frame gives you a better sense of where you expect the populations to potentially equilibrate from applying a particular harvest rule.

After that, even the counsel was thrown off balance. A painful silence ensued as she was rummaging through her notes searching for a follow-up question which would allow her to temporarily conceal her newly-found state of cluelessness. She came up with this:

Q: the model assumes – correct me if I'm wrong – that the past history of productivity in the past [sic] will be predictive of the behaviour of stocks in the future?

Oh boy, that was deep. She was asking him in substance: are you telling us that your model makes predictions about the future based on the past? Well yes Ma’am, that’s usually what models do.

But, being fundamentally a nice guy who is deeply passionate about his subject matter, Jeff Grout did not seem to notice the clumsiness of the counsel’s question, and responded almost enthusiastically with the following:

A: The model itself uses the information from spawning and recruitment and distribution of the annual variations about that. In the initial formulations of the model, we were just looking at the historical spawner recruitment data, but in recent revisions to the model we have added elements allowing us to look at different productivity scenarios moving forward into the future – by that I mean you can look at a continuing decrease in recruits per spawner eventually, or maybe something that goes back to historical patterns, break it even, put in your own series of productivity in the future to see what the potential........................

I lost what he was saying after that because a fellow activist sitting next to me came to my brain’s rescue by exclaiming rather loudly into my ear “He lives in a cloud, that guy!” I stopped taking notes after that and went instead into meditation mode, as the esoteric blathering continued like a shopping mall music in the background.

Not a single time in his two last long cryptic responses did Mr. Grout use the word ‘salmon’ or ‘fish’, I realized. That can be seen as a problem, given that his job title is – after all – “Regional Resource Manager, Salmon”. Instead, he used terms such as ‘populations’, ‘spawners’, ‘harvests’, ‘recruits’ to abstract this magnificent animal into a set of mathematical variables which could fit in his equations. Mr. Grout does not deal with the living salmon, but instead with conceptual entities, statistical mind games, complex sudokus for PhDs which he is being paid by the taxpayer to resolve. The word ‘salmon’ in his job title? A mere embarrassment on his business cards, an awkward reminder that he, too, is after all a carbon-based organism receiving justification for his paycheck from another carbon-based organism which he only seldom gets to see or think about. The question for Mr. Grout I desperately wanted to whisper in the ear of the Commission counsel to help her get back on her feet was: “Mr. Grout, how many times have you left your corner office this year to go out in the field and see the Fraser sockeye in their natural environment?”

When the First Nations leaders testified before the Commission last month, one of their main complaints about DFO was the fact that this agency claimed to hold exclusive rights over salmon knowledge, that it dismissed aboriginal traditional knowledge as pseudo-science, and that it used it to assert a complete monopoly over decision making, in contradiction with its mandate to co-manage the resource with the First Nations. This resulted in a top-down neocolonial attitude whereby, according to the witnesses, DFO would unilaterally say to aboriginal communities: we in our wisdom have determined your fishing quotas for the year. Here they are. Sign here, or you don’t get to fish.

Sitting in that room with Mr. Grout, I was definitely getting their point. A bureaucracy such as DFO really does epitomize the classical division between manual and mental labor, which in Marxist theory is one of the main instruments of control of the ruling class over the rest of society. We know. You don’t. And so we decide and we organize, and you get to execute according to our decisions and modes of organization. Here are your quotas. Go fish. Don’t ask how we got that number, our models wouldn’t mean anything to you anyways.

You don’t believe that this is actually how DFO thinks? Well then, listen to the following exchange between the Commissioner’s counsel and DFO’s Area Director for the BC Interior, Barry Rosenberger, who testified at the Commission on that same day:

Q: Do you think the stakeholder groups have the capacity to understand the issues that are presented to them for decision and feedback, including some of the technical work that we just touched on today?

A: The level of technical capacity for some of the groups varies for sure. Some of the groups definitely want and expect the Department to have that capacity, to bring them that information and them to be able to give input based on that. Other groups are trying to have people that understand all the models at the same level that we do.

Q: Do you think the stakeholder groups need to understand the technical workings of the forecasting models in order to provide meaningful input?

A: I do not. I don't know the technical workings of those models myself. We can't all have PhDs and all have the same expertise, and we're all going to be geneticists and modelers and whatever. We have to be able to get that information from someone. I don't think it's effective to expect that they are going to have 10 or 15 people developing their own models and having discussions around this.

The strange thing, as one of my fellow activists with a scientific background remarked, is that the key principles behind DFO’s insufferably complex forecasting models are rather self-evident – actually, even lame – if one would accomplish (as she did) the superhuman effort of cutting through the bullshit. Her aha! moment came when Mr. Grout commented on a very complicated graph representing three possible return trends for the Fraser sockeye. All they are doing here, she told me afterwards, is take three basic scenarios – optimistic, middle of the road, and pessimistic – and provide the agency’s decision makers with those three options to pick from.

What my friend was pointing to was the fact that the very language used by DFO bureaucrats was the main cause of their work's complexity, rather than the work itself. And why should we be surprised? We all know since Carl Sagan and Stephen Hawking that the basic concepts of science – even special relativity, of all topics – can indeed be explained to the layperson if one chooses to leave a few details out. When one, instead, overburdens the listener with tedious jargon and unimportant minutiae, as Mr. Grout did for a whole day at the Cohen Commission, one is performing (unknowingly, I’ll grant him that) a political act which consists of locking knowledge away from people by using an encrypted code.

There is another institution which has performed this type of encryption of knowledge for millennia in an effort to subjugate and control the masses, using the division of labor between intellectual and manual tasks to achieve its purpose – and that is religion. We know, you don’t. Only we the priests can speak to the gods, that is why you must work for us and provide us with regular offerings in wheat and gold. The primary tool of religious power over the masses is ignorance. And ignorance can indeed be manufactured, by transforming the simple into the complicated, by turning what everyone once knew into what only the initiated can now comprehend. Like the priests of the world's great religions, DFO bureaucrats have invented and honed over time an esoteric new language and set of rituals understandable only to them, which they call knowledge and use to hold power over what they call the "stakeholders".

The great irony in this knowledge power trip embodied by Messrs. Grout and Rosenberger is that DFO's forecasting models of the Fraser sockeye runs have been consistently and massively wrong over the past decade. For example in 2009, they predicted that 10 million salmon would return, when barely a million did. In 2010, determined not to make the same mistake again, they announced a ridiculously wide range of between 7 and 18 million salmon – and missed the mark again, as 32 million sockeye came rushing up the Fraser. In effect, DFO’s fancy and expensive models, which only they can (and claim they should) understand, are useless.

One of the core demands of the First Nations leaders when they testified before Justice Cohen was this: we need funding to hire our own biologists, so we can integrate modern science with our traditional knowledge of the salmon. They were spot on. This is indeed where the battlefront lies. Break the division between mental and manual labor maintained by DFO's self-serving bureaucracy and the legions of “independent” scientists who gravitate around it. Reclaim the knowledge from those who took it away and who are keeping it vaulted in a custom-designed apparatus of scientific-bureaucratic hocus pocus. Indigenous people have long known what we settlers are only now awakening to – that on the Fraser, like everywhere else, knowledge is power.