Wine democracy, part II: Crowd-sourcing

If one way to make wine more democratic is to make wine writing more “accessible,” another is crowd-sourcing, asking “the consumer” what they want and finding ways to make it for them. Washington State’s Columbia Crest billed the “Crowdsourced Cabernet” it released in June of this year as “the first wine to be crowdsourced all the way from the vineyard to the bottle” via community input solicited online and, of course, filtered through one of their staff winemakers. I’ve yet to find one of the resulting 12,000 bottles in Edinburgh and can’t comment on the result, though there’s only so far you can go wrong with a $30 Horse Heaven Hills cabernet.

Taking crowdsourcing in a different direction, Brock University and Ontario Grape and Wine Research announced a new initiative this past summer to increase Ontario red wine sales by monitoring tannins to help winemakers produce the “rich and robust” reds that “the consumer” wants.

I have to put “the consumer” in quotes for the same reason that I, as a responsible scholar of science communication, have to put “the public” in quotes. In both cases, there’s no such thing. There are multiple publics, and multiple consumers, and anyone talking about them in singular form is either imagining a more specific group of people or being horribly vague.

That problem – the problem that “the public” consists of all manner of different people – is at the heart of the problem with crowd-sourcing. Crowd-sourcing calls on averages: the strategy takes a whole bunch of individual views and homogenizes them into a single outcome. Crowd-sourcing makes “the consumer” into a single group that votes to produce a single outcome that is then supposed to make most people happy enough most of the time. Crowd-sourcing imagines that the customer is always right, displaces passion, and erases diversity.

The customer is always right is wrong: That appears to be fairly common business knowledge, at least in the post-Jobsian era in which we’ve all been deeply saturated with i-products we never knew we wanted. Customers don’t always know what they want. For starters, their professional expertise doesn’t lie in arriving at new commercially viable solutions to daily problems. And customers certainly don’t know what manner of new and previously unimagined product they’ll buy when presented with the option to do so. Being asked a question about what you would like is different than being asked whether you do like something actually in a glass in front of you.

Passion is what makes wine: Passion is one of my least favorite words. It crops up on resumes in unlikely places, has been co-opted by business jargon in the service of banal and insulting sales pitches, and is pulled into service as a catch-all for people who haven’t thought deeply enough about what motivates them. Passion is also an enormous part of what makes wine, though I could just as easily call it pigheadedness. One person or a few people in collusion have an idea of something they’d really like to see happen because it would make them happy. They pursue it in the interest of making themselves happy and – lo and behold – other people are made happy by some of the same things, and that kind of satisfaction is contagious. Replace one person’s idiosyncratic passion-driven pipe dream with too much market research and a game of averages and you end up in a world of desk jobs. Most of us did not become interested in wine because we were looking for more desk jobs.

Diversity is a good thing: Consumers want “ripe, rich, rounded red wines?” I’m certain that some of them do. Heck, sometimes I enjoy the robustness of a big Horse Heaven Hills cabernet alongside the great diversity of other red wines I can find in eastern Washington. Crowd-sourcing as a fun marketing tool is one thing for an individual big-brand winery like Columbia Crest. It’s another thing to apply a general idea of what “the consumer” wants as a tool to guide the production of an entire region. Do we really want to encourage wine production to be less diverse, even if doing so increases sales? It’s an open question with many possible answers. But I’m skeptical that anyone, even the large producers making inoffensively homogeneous wine, wins by making wine more the same.

All of that said, I’m skeptical, but not worried. The best thing about the wine community isn’t that it’s democratic so much as that it’s a free state. Wine crowdsourcers can do their thing; plenty of more interesting wine will still be around for those of us who’d rather not follow the crowd.

Effects of grapevine leafroll disease on wine quality (and when is a disease a disease?)

Gut reaction: Viruses cause disease. Disease is bad. Viruses are bad.

Gut reaction muted by a lot of recent genetics research: Viral DNA seems to be embedded in genomes all over the place. We’re not sure why a lot of it is there, or stayed there, or what it does while its there. Some viruses cause disease. Some don’t. Viruses are complex, and we probably don’t know the half of it yet.

A name like “grapevine leafroll-associated virus” gets you thinking about negative consequences. Rolled leaves don’t collect light efficiently, which means that they won’t contribute to the plant’s photosynthetic metabolism efficiently, which means that the plant may be malnourished, grow slowly, and/or not have enough energy to ripen fruit. Rolled leaves are bad. A virus that’s associated with rolled leaves is bad. But the virus is only associated, not causative. Some viruses in this general family of leafroll-associatedness aren’t associated with vine symptoms. And infected vines only show symptoms post-veraison (the stage of ripening at which grapes change color), even though they carry the virus in detectable quantities year-round.

Ergo, a group of vine and wine scientists headquartered in eastern Washington state designed an experiment to ask (published in PLOSOne, and therefore open-access to everyone): do grapes from vines with grapevine leafroll disease, and carrying one of these viruses (GLRaV-3), lag behind their undiseased counterparts throughout ripening, or only when vines show symptoms? Being particularly conscientious*, they also improved on existing studies of grapevine leafroll disease by collecting data for three consecutive years from a commercial vineyard, sampling grapes throughout the season but also harvesting grapes at the typical time and making wine from diseased and undiseased pairs, and subjecting those wines to (limited) chemical and sensory analysis. They also used own-rooted rather than grafted vines, which eliminates some potentially confounding variables.

Their conclusions, after collecting data over the 2009, 2010, and 2011 growing seasons:

  • Grapevine leafroll disease decreases vigor (as measured by cane pruning weights) and fruit yield in own-rooted Merlot vines in Western Washington.
  • Grapes from diseased/infected vines have lower total soluble solids (TSS) and higher titratable acidities (TA) (and, to a less dramatic degree, lower anthocyanin concentrations) than grapes from undiseased/uninfected vines, but only after vines begin displaying symptoms post-veraison.
  • Wines made from diseased grapes were browner and less intensely colored, earthier and less fruity, and more astringent compared with their undiseased counterparts.
  • However, panelists only correctly distinguished diseased from undiseased wines when served side-by-side in black glasses, removing the notable color differences from consideration and forcing them to differentiate on smell and taste alone.
  • Soluble solids, TA, and pH were all more dramatically affected than anthocyanins in diseased vines, which reflects the decoupling of anthocyanin development and sugar accumulation that happens late in ripening during which environmental conditions play heavy in anthocyanin development.

These conclusions probably do more for plant scientists than for commercial growers: data from one Merlot vineyard near Prosser can’t be precisely extrapolated to you, wherever you are, and thresholds for usable fruit are always a matter of context (the authors note that future studies should document the effects of grapevine leafroll disease on specific sensorily-important compounds). The study does add data points to a collection of statistically robust data that might help large companies make judgments about what they can include in their generic red blends before pH or some other parameter becomes a problem. But maybe the most interesting line of thinking here has to do with the nature of disease, and of relationships between viruses and diseases and symptoms. Do vines have leafroll disease before they exhibit symptoms? Where do we want to draw lines between normal or acceptable variation and disease symptoms? If a vine looks sad but makes grapes that make wine indistinguishable from happy-vine wine, and if genetic testing says that the plant also happens to have a virus, does that mean that the vine has a disease, or is it healthy?

Disease” can mean something different to the plant pathologist who looks at a vine, a geneticist who looks at the DNA of a vine, a commercial grower who looks at the fruit of the vine or a winemaker who looks at the juice it makes. That vine may be infected with viruses. Is the virus bad?

 

*Full disclosure: I know and think highly of several of the scientists on this team.