When I was young, I would sometimes go lake fishing with my father and uncle in the Canadian wilderness. Our boat was equipped with sonar that reported water depths on a big-pixel screen. We would look at sketchy patterns in the data to make inferences about the underwater topography. The goal was to find craggy terrain (cliffs, hills, holes, and rocky outcrops) where fish congregate to eat or be eaten. On rare occasions, the sonar would show schools of fish passing underneath the boat. This wasn’t the only data we used, however. We often had access to low-resolution maps that charted water depths within lakes. Some of these maps even contained bits of folk wisdom in the margins and served as decoration on the walls of lodges and cottages. There was also the experience of local fishers, who remembered the places where they had success. Some would trade tips about where the good spots are located. All of these accounts and records are forms of empirical data of varying degrees of accuracy, precision, reliability, and timeliness. (By empirical, I simply mean careful human observation.) My father and uncle would take all this data into account to determine where to drop the fishing lines.

I also remember hearing stories about how fishing was so much easier back in the old days. Fish stocks had declined over the decades. Some lakes were deemed to be over-fished or not very bountiful in the first place. Such spots were avoided. All of these recollections made me wonder what these lakes were like in their original, pristine form … or at least in the days before sport fishing. I wondered because I was a pathetic fisherman who yearned for abundant, dumb prey.

I was reminded of this while reading about historical ecologist Ted Ames in the latest edition of The New Yorker.1 Ames recently won a “genius grant” from the MacArthur Foundation, a philanthropic trust. Ames spends most of his time as a lobster fisherman in Maine. Given the collapse of the cod fishery off the coast of New England, he wondered what cod stocks were like in the past, especially the location of defunct spawning grounds. This might help efforts to repopulate the fishery. The problem is, as the man himself puts it: “You’re tracking things that you can’t see, that are moving around in the past …” (p.61)

Methods used by U.S. government officials were not working, according to the author of the article, Alec Wilkinson. The officials were dropping nets at random within an extremely large coastal zone. If they found increases in their catch, they assumed that fish stocks had rebounded and then authorized more fishing. The problem with this is that the fish are distributed in identifiable patterns, such as regular spawning and foraging routes. If the small random sample hit one of these concentrations, government officials would make an incorrect inference about rebounding fish stocks. The error rates for the samples were enormous. This method also didn’t offer many insights about actual fish behaviour and no insights whatsoever about stocks in the distant past.

Ames decided to look at oral histories. He mapped fish concentrations and migration routes in the past using accounts from the older fishers. Fishers could point to places where they used to catch lots of fish. Skilled fishers could do that accurately because they know the topography of the bottom of the ocean in detail, including where old ship wrecks are. These days, elaborate sonar equipment and detailed topographic surveys are available; back then, a watch and a compass were the only guides. These mental records were required because the fishers had to remember where the best fishing spots are, as well as what trouble-spots to avoid (places where their fishing nets could get damaged). Now that the fishery is dead, the fishers are willing to be candid: they no longer have a stake in withholding information but do have a stake in replenishment efforts. Ames would only plot a report on his map if it was corroborated by another fisher, thus adding some validation to the data collection. Based on other evidence about spawning habits—cod return to where they were born in order to reproduce—Ames was able to find many of the old spawning grounds. In sum, he created a picture of the past using empirical data found within oral histories.

Canadian government scientists have used a similar technique with the oral histories of aboriginal elders.2 These cases can be found within a series of documentary vignettes (found here). For example, scientists have been mapping the fish habitats in the Bras d’Or lake system of Cape Breton, Nova Scotia. This is done with sophisticated electronic sensors, left in the lake for a year, that record enormous amounts of data about various lake conditions (e.g., water temperature). This provides a picture over a relatively brief period of time. In order to get a better sense of how the ecosystem has evolved, and how humans have influenced the ecosystem over time, other means are required. The researchers chose to extract empirical data from the oral histories of the Mi’Kmaq people in order to create additional layers on the map. For instance, a layer based on old hunting and fishing spots was created. This adds to the usefulness of the map and helps scientists with interpreting the other data.

Of course, these methods have weaknesses. Human memories degrade over time. Oral histories do not always contain accurate information. Vague accounts must be interpreted. Aboriginal oral histories often combine empirical observations with religious interpretations, which must be disentangled. Some assumptions about fishing skill and technological development are made. Oral histories can change in subtle ways as accounts are passed from person to person. Some embellishment is expected: old fish stories are, after all, old fish stories; there are countless folk-sayings about how lying is part of the act and ritual of fishing. Moreover, fishers observe what matters to them, which may not be the same things that matters to ecologists. Nonetheless, this strikes me as a very clever way of solving (or perhaps partially solving) a tricky research problem.

Another clever approach involves using written records to reconstruct the past. A palaeo-oceanographer named Glenn Jones pulled data out of roughly 40,000 American seafood-restaurant menus dating back to the 1850s.3 This provides evidence of what fishers were catching for the consumer market. Changes in price often indicate changes in supply, such as the exhaustion of local stocks. Of course, price fluctuations can also reflect a change in cost, demand, or distribution system. To partially control for such factors, Jones compared retail prices to 50 years worth of available wholesale price data. He found that both prices changed more-or-less in tandem. He adjusted for inflation. He accounted for declines in demand due to income changes (e.g., the Great Depression). Jones also looked at records of changing consumer taste (e.g., the growth of lobster as a luxury good as stocks declined), a finding that was partially discovered by grading the restaurants as a restaurant guide would (a five-star rating for a fancy restaurant and a one-star rating for a dive, based on historical records about each restaurant and information in each menu).

Again, Jones is only providing a partial picture of the past. That said, the menus are empirical data that can help researchers reconstruct the past, especially when combined with the knowledge of experienced fishers. Fishers have quite a stockpile of data, even though they don’t think of it as such. I wouldn’t call any of this “scientific” in the true sense of the word, unlike most others who have cited these projects. Yet, drawing lessons from such data sources, while accounting for potential biases, is certainly better than just guessing about the past. As Mark Twain once quipped: “Do not tell fish stories where the people know you; but particularly, don’t tell them where they know the fish.”

By Peter Stoyko

NOTES

1. Alec Wilkenson, “The Lobsterman: How Ted Ames Turned Oral History into Science,” The New Yorker, July 31, 2006, pp. 56-65.

2. This project was administered by Fisheries & Oceans Canada in collaboration with the Eskasoni Fish and Wildlife Commission. The vignettes were produced by Telefilm Canada (2004).

3. For an overview of Jones’ findings, see: COML History of Marine Animal Populations Project, “Restaurant Seafood Prices Since 1950s Help Plot Marine Harvests Through History,” News Release, ‘Oceans Past’ Open Science Conference, October 24-27, 2005, Kolding, Denmark. This news release is available here.