JOY had always been an idiosyncratic operation
January 5, 2019 12:23 AM   Subscribe

In 2009, Cornell food psychologist Brian Wansink published "The Joy of Cooking Too Much", where "...[he] had examined the cookbook’s recipes in multiple 'Joy' editions, beginning with the 1936 version, and determined that their calorie counts had increased over time by an average of forty-four per cent." In 2018, Buzzfeed News published an exposé on p-hacking at the Cornell Food and Brand Lab, based on work by Wansink's critics, and in light of that, "Joy" decided to investigate more closely.
posted by frimble (42 comments total) 30 users marked this as a favorite
 
Why do useless dishonest jerks like Wansink still have jobs?

It used to be that if I saw a splashy headline about some scientific discovery, especially in the areas of nutrition or psychology, I assumed that there were nuances and limitations in the paper that the journalism had failed to capture.

Now, I simply (in those areas) don't believe it at all until it's been replicated. I assume it's been jazzed up in some fundamentally shady way.

While cynical, I'm a science-friendly person. I wasn't naturally inclined to think this way. Yet here I am. So many studies have proved unreplicable, so many professors appear to have outright misrepresented their work, and yet so few consequences have redounded to anyone.
posted by praemunire at 3:03 AM on January 5 [29 favorites]


Ugh, thanks, Wansink, for giving people an actual reason to distrust science.
posted by the_blizz at 3:22 AM on January 5 [10 favorites]


Heh, I came in to say almost word for word what praemunire said. Fake or shoody science and bad reporting around it is all incredibly frustrating and can have real consequence. It shouldn't be tolerated.
posted by gusottertrout at 3:25 AM on January 5 [1 favorite]


praemunire: Why do useless dishonest jerks like Wansink still have jobs?

He has announced his retirement, which will take effect in June.
posted by James Scott-Brown at 3:49 AM on January 5 [2 favorites]


Why do useless dishonest jerks like Wansink still have jobs?

In the specific case of Wansink, so he can assist Cornell in its ongoing review of his prior research.

Cornell Prof. Brian Wansink Resigns After Academic Misconduct Investigation
Prof. Brian Wansink has tendered his resignation following the conclusion of an investigation of misconduct that was conducted over more than a year, according to a statement from Provost Michael Kotlikoff. Wansink will not be permitted to teach or research at the University, and will retire at the end of the academic year.

A faculty committee concluded the investigation, and found that “Professor Wansink committed academic misconduct in his research and scholarship, including misreporting of research data, problematic statistical techniques, failure to properly document and preserve research results, and inappropriate authorship.”
posted by zamboni at 3:50 AM on January 5 [6 favorites]


This is what happens when you create a publish-or-perish environment in academia, and then refuse to publish negative studies unless they have some sort of "spin." I've seen this "slicing and dicing" approach in labs ranging from bench/animal research to population health. It's one of the reasons I left Science -- I don't want to spend years of my life and thousands of taxpayer dollars only to end up with no job security because I failed to meet a publication metric.

Not to excuse what Wansink has done, but he is the product of a broken system.
posted by basalganglia at 3:53 AM on January 5 [37 favorites]


Came to say what James Scott-Brown and zamboni already have: Wansick has been forced out at Cornell. Yes, it takes effect at the end of the academic year, but he's done. For that matter, seventeen of his papers have been retracted. He's done.

It's cases like this that get me to reaffirm my belief that if you're going to do science, you need to be making the data available. Particularly, I would argue, in more applied research areas, where there is (historically) less of a trend towards openness. Does this kind of crap happen? Yep. Probably more often than we really want. What's the best disinfectant? Being able to dive into the materials, methods and data and say "hmm, something smells."
posted by Making You Bored For Science at 4:50 AM on January 5 [10 favorites]


One point that remains consistent across virtually every nutrition and health recommendation is that eating home-cooked meals prepared with fresh ingredients correlates with better health.

Isn't this also the kind of recommendation that just happens to dovetail with the current cultural zeitgeist and could easily be based on Wansink-style research? Like a recommendation to take more vitamin pills in the technology-positive 1950's?

Of all Wansink's work, this result actually seemed the most believable; I had always assumed that portion sizes in cookbooks were indeed increasing because many of them had previously been based on the idea that you have time to make a salad, first course, three sides, etc.
posted by Ralston McTodd at 5:04 AM on January 5 [2 favorites]


One of his better papers won a 2007 Ig Nobel in Nutrition.
posted by ikalliom at 5:09 AM on January 5 [1 favorite]


I used to share Wansink’s book liberally and actually got rid of my giant 1990s plates in part because of it. I kind of feel a burn at his misuse of data, grad students, authority, funding, and the word science.

However, as the Buzzfeed article says if I’m remembering it right, it is, in fact, the commitment of a scientist to examining results which led to Wansink’s disgrace, as well as the idea that you can and should attempt to reproduce things (math, experiments.)

So while I do feel like we’re in a really low point for science in some fields, Science herself is carrying the day.

Also my edition of Joy is the stained manual I pull out over and over whenever I need to deal with some mysterious package from our ag share bin, although it has failed me with goat.
posted by warriorqueen at 5:27 AM on January 5 [9 favorites]


One point that remains consistent across virtually every nutrition and health recommendation is that eating home-cooked meals prepared with fresh ingredients correlates with better health.

While I enjoyed this article and deplore Wansink, the writer mistakenly positions sixties JoC writers as unusually prescient. There's a history, going at least back to the thirties, of mass market cookbooks that advance precisely this hypothesis (and statement that itself assumes you're disregarding earlier proto-hippie/alternative culture books going back to the late 19th century.) There's at least one mefite who's an actual culinary history scholar who could lay this all out in detail, but it's a hobby for me and I'm thinking of both MK Fisher's early work and Adele Davis's Let's Cook It Right, of which I have a forties edition.

Not that I don't love the Joy of Cooking, although TBH I prefer my sixties copy because you can get contemporary recipes anywhere. It's a very solid, pleasant book and I'm glad to see that they were right.

This reminds me of that post on the blue a couple of weeks ago about the German journalist who made up all that stuff about Fergus Falls MN, and how FF people did their own research to show that he'd lied. It's frustrating that people with authority can just lie and get traction. I'm excited by the gumption of ordinary people doing or arranging their own research, but it shouldn't be the case that ordinary people have to go to war with funded "experts".

In re Wansink: It's not just publish-or-perish. I have known some scientists who publish a lot, and some scientists who really do have to scramble in cut-throat situations, and who don't lie. It's that science - especially big, mediagenic academic science at a big institution - rewards arrogant jerks, and the big, arrogant jerks tend to end up with contracts that mean it's virtually impossible to fire them for actual misconduct and therefore they just ease on into "retirement". (I've actually seen one recent instance of this that genuinely shocked me given the high visibility of the misconduct, some of which I witnessed.)

Wansink seems to me very much like every other high-profile white male moralizing academic that a big university wants to use to get media attention. Naturally he has a counter-intuitive hypothesis, not about Paula Deen recipes or something that wouldn't be sufficiently splashy, but about Joy of Cooking. Naturally he knows absolutely zippo about cooking history (boring humanities garbage that concerns itself with women) and so doesn't know or doesn't care to know about how to evaluate a cookbook. He's That Guy, and big schools select for That Guy.
posted by Frowner at 5:38 AM on January 5 [42 favorites]


It makes me happy to see Helen Rosner's work shared here. She is an awesome human being and writer, and I can think of no better person to write an article taking a jerk like Wansink to task.

Not that I don't love the Joy of Cooking, although TBH I prefer my sixties copy because you can get contemporary recipes anywhere. It's a very solid, pleasant book and I'm glad to see that they were right.

Growing up, the chefs in my life implored me to never buy any other copy than the 1964 edition (or prior). I didn't realize how critical and spot on their advice was until I read this today.
posted by nightrecordings at 5:56 AM on January 5 [5 favorites]


Here's something I never knew existed. There's a "Retraction Watch" database...
posted by mikelieman at 5:59 AM on January 5 [2 favorites]


In defense of Joy

Also my edition of Joy is the stained manual I pull out over and over whenever I need to deal with some mysterious package from our ag share bin, although it has failed me with goat.

Oh gosh I checked and my early 70's edition references Lamb, Mutton, Deer, Moose, Elk, Squirrel, Woodchuck, Beaver, Raccoon, Wild Boar and yes, even Peccary there seems to be no reference to Goat? I'll have to ask my sister that got the real edition (40's) if there's anything earlier. I suspect it's considered essentially like mutton but does seem to be an oversight, stunned folks, just stunned.
posted by sammyo at 6:09 AM on January 5 [4 favorites]


I worked in food photography studios in the 90's. I remember when the "New Coke" version of JoC was released. A food stylist I worked with told me I should have a copy, but to make sure it was the pre '97 edition. When I expressed surprise that food stylists kept JoC on hand, she told me something like "well, Joy won't have the perfect recipe for cake, but it's very helpful if you need to figure out what cake is."
posted by Drab_Parts at 6:14 AM on January 5 [7 favorites]


and actually got rid of my giant 1990s plates in part because of it

We did as well. At this point I prefer the smaller plates just because that is what I am used to now, but it's mildly irritating that it was fraudulent science that made me consider the change in the first place. The plates thing is just a silly sidebar; other parts of his work had serious, real-world implications and I wonder how much harm he caused.
posted by Dip Flash at 6:15 AM on January 5 [6 favorites]


I just checked and I do in fact own the 1997 edition. It was a college graduation gift, and I remember my mom saying it was too bad it wasn't an older edition. I don't use the actual recipes that much, except for baking because internet baking recipes are so often garbage, but it is my first stop when I'm dealing with an unfamiliar fruit, vegetable, or cut of meat. It is the very best reference book in our kitchen. And unlike the earlier editions, it does indeed have a section on goat.
posted by hydropsyche at 6:38 AM on January 5 [1 favorite]


I paid $150 to get our 1964 edition professionally rebound. The text block was in ok shape (with stains that indicate which recipes my parents made most often) but the binding and covers were shot. I digitally re-created the endpapers and the cover cloth on laminated lightweight canvas. It's as everyone says: maybe you don't make the exact recipes in the book, but it is the ur-cookbook from which a great deal of kitchen knowledge springs.
posted by seanmpuckett at 6:43 AM on January 5 [14 favorites]


In re Wansink: It's not just publish-or-perish. I have known some scientists who publish a lot, and some scientists who really do have to scramble in cut-throat situations, and who don't lie. It's that science - especially big, mediagenic academic science at a big institution - rewards arrogant jerks, and the big, arrogant jerks tend to end up with contracts that mean it's virtually impossible to fire them for actual misconduct and therefore they just ease on into "retirement". (I've actually seen one recent instance of this that genuinely shocked me given the high visibility of the misconduct, some of which I witnessed.)

I think some entire fields are more prone to bad research and misconduct than others. Anecdotally, I said this in a comment at a big cross-disciplinary conference about research funding, and I've never had so many people from different fields come up to me during the coffee-break, confirming my assumption with their own examples.
I didn't know about Wansink, but nutrition is one of the fields on my list. That doesn't mean there is no good research in nutrition, it just means it's a field where I will double-check every detail in an application if I'm on a board.
posted by mumimor at 6:45 AM on January 5 [6 favorites]


I have two copies of Joy - mine which is from 1975 and my mother's which was the first edition after 1943 according to the forward but the publication date page is long gone - both copies held together with tape and thoroughly stained and tattered. Great reference books - I learned a lot reading them in years past and still find mine useful to recheck techniques. No goat recipes in either to my surprise.

Have often thought that we'd be far better off if scientists were rewarded for publishing negative results. Those are often more significant, useful and would help keep things more honest. The pressure to publish only positive results distorts things tremendously. I am not a scientist but am one of the few non-scientists in my family so hear about this a lot.
posted by leslies at 6:55 AM on January 5 [1 favorite]


and actually got rid of my giant 1990s plates in part because of it

We did as well! I'm 100% sure I never read the Wansink's book about the horrors of Big Plates. That demonstrates that his "findings" have sifted on down to people like me, those that didn't really pay attention to nutrition science and food research.

In my previous library job I was responsible for a cookbook collection of over 10,000 volumes. Our (the) Joy(s) of Cooking are very, very heavily used. So much so that we had multiple copies in Reference, just so that we could keep the older ones in good condition...and, of course, so they didn't walk away and find new homes in someone's kitchen. I also had extra, non-public copies in my office just in case something happened to one of our public copies and we needed to replace it.

It was my go-to Reference for when a patron would call in looking for a recipe...some of my favorite Reference calls were things like "I need to make deviled eggs for 50 by 6:00 tonight. Do you have a recipe I can use?". I'd say "Yes! I have the Joy of Cooking right here..." and the caller would inevitably say something like "oh good." From my recipe-giving experience, the Joy of Cooking is universally accepted as the Ur-Cookbook: trusted and loved.
posted by Gray Duck at 6:59 AM on January 5 [19 favorites]


My 2006 edition has a single recipe for "Goat, Jamaican curried." There is also "Roast Mountain Goat or Bighorn Sheep" in the game section (right above the Wild Boar and across the page from Braised Bear); I can only assume they were trying to capture the Ron Swanson crowd with those.
posted by basalganglia at 7:26 AM on January 5 [2 favorites]


I have multiple JOC, inherited as women in my family have shuffled off the mortal coil. The 40s and 60s versions are my faves, with instructions on how to carve up a deer, cow, and other livestock, explanations of chemistry, and little notes like, if you don’t have X, use Y. Three generations of cooks in my family have written notes in the grease stained margins. I consider the Early JOC to be the American larousse gastronomique
posted by SecretAgentSockpuppet at 7:57 AM on January 5 [7 favorites]


What I take from this story and Klaas Relotius (the German journalist who faked substantial parts of his stories) is that, while there may be some checking in general, you'll get caught faster (not very fast) if you stomp on real people's toes.
posted by Nancy Lebovitz at 7:59 AM on January 5 [2 favorites]


it is the ur-cookbook from which a great deal of kitchen knowledge springs.

In Chinese cookery in the US, there is another one of those: The Thousand Recipe Chinese Cookbook, published in 1966 and written by Gloria Bley Miller. My (Chinese-American) father swore by it.

Regarding Wansink and those "scientists" like him, some folks are real scientists who start out doing real science, get some celebrity by winning awards, and THEN start making shit up so they'll stay in the public eye (I knew Melvin Calvin, a nobel prize winner in biochemistry, in an unrelated field - he was Dad's mentor - and his science about vegetable metabolism was good until he started doing self-experimentation in cholesterol and pumping/pimping that for the press). There are also scientists who are a little crackpotty to begin with, like Linus Pauling and his Vitamin C megadosing. A lot of older scientists during his time were starting to get the fear of mortality as they got older, so they started coming up with and believing anything (like anti-oxidants, which showed a lot of initial progress until the awesome and ineffable process of poorly understood and individual human metabolism and enzymatic reactions got in the way) that might possibly give them an edge against all those years of smoking (whose hazards they helped suppress earlier in life), for example.

Also it's actually possible for some do-badders to do it out of cluelessness. I don't think that's the case for Wanksink never having heard of p-hacking. That seems super disingenuous. But educational standards about statistics, experimental design, and analytics within scientific degree programs are pretty hit and miss. I got a lot of extracirricular education from my Dad, a scientist who'd seen other scientists go bad or whackadoodle. And I had luck enough to get a scientific degree through a program that really pushed hard on scientific veracity, the scientific method, the philosophy of scientific experimental design and of the scientific method, and on being super careful with data so as not to compromise it during analytics. But not everyone has that background. Some people just learn the first layer of doing science and don't go into the whys and hows as thoroughly as I got. Some programs aren't so careful and they turn out scientists who are not careful.

And a lot of it has to do with timing. My (incomplete) postgraduate scientific education happened with instruments that were just then getting a shakedown - folks had been experimenting with the Atomic Force Microscope and post-experiment data analytics and getting really wild and unsupportable results - basically seeing what they wanted to see - so I got some training in that too, because of the shakedowns in that part of science at that time.

I think, unfortunately, there are a lot of very human reasons that scientists will turn, and compromise the academic and scientific integrity and philosophical lessons and reasons for doing good science. Some of it boils down to fear, or want of celebrity, or intimidation or greed. There are just so many reasons to go dark, and so few rewards for doing the right thing. I think if we (all we people) treated each other better there would be less overall stress and less need to go above and beyond and compromise ethics in search of things we desire.
posted by kalessin at 8:27 AM on January 5 [11 favorites]


Also I don't mean this commentary as character or personality assassinations of these two scientists I named. Melvin Calvin was kind of nice sometimes, in private, but kind of a dick celebrity-type sort of full of himself a lot of the time, especially after the Nobel prize. Pauling was actually quite a sweet guy who had a tendency to go on a rant about possibilities. A real dreamer, but a nice guy.
posted by kalessin at 8:29 AM on January 5 [1 favorite]


I never thought to check until just now but apparently I have the 1975 version which has been my go-to for anything unfamiliar or battle-tested/american since like whenever.

Pretty sure I have a (softcover?) older version somewhere, I remember not wanting to use it too much because the binding was disintegrating.

Is there somewhere I can look to see the changes from version to version? Like a version controlled JoC? What about the 1965 version makes it the definitive version?
posted by pagrus at 10:59 AM on January 5


When I was just starting out learning to cook (beyond basic things like grilled-cheese sandwiches, scrambled eggs, Rice-a-Roni, etc.) the first recipes I learned were from my mom's mid-70's, maybe? JoC. I also recall that early on I came up with my own tomato sauce recipe cobbled together from bits of the 3 or 4 variations in that book (I've since gravitated to a simpler recipe for most of my pasta dishes that doesn't take as long to make, but I still use my original one when I make lasagna because it's so rich and deeply flavorful and satisfying in that heavier dish).

Later I got my own copy, an 80's or 90's edition, which somewhere along the line got swapped for a 1964 edition - I have no idea how or why that happened - that I still have today. I haven't pulled it off the shelf in ages but this thread has inspired me to do a browse-through. I make no predictions or promises about the portion size of whatever I decide to make from it...
posted by Greg_Ace at 11:01 AM on January 5


Tangentially: the Buzzfeed piece is full of puns, which is good or bad depending on one's perspective.
posted by Sheydem-tants at 11:50 AM on January 5 [1 favorite]


Brian Wansink is a charlatan who has fouled up the research mission of academia. Brian Wansink anagrams to "wanks in brain."
posted by duffell at 1:30 PM on January 5 [1 favorite]


Wansink said that he’d decided to analyze “Joy” because he was looking for culprits in the obesity epidemic beyond fast food and other unhealthy restaurant cooking.

Related to the sugar FPP a few weeks ago and the world being what it is, I wonder who was bankrolling him (directly or indirectly) in the corporate agriculture sector.
posted by MillMan at 2:32 PM on January 5 [4 favorites]


I just had a chance to check -- the JOC on my shelf appears to be the 1964 edition, passed down in the family. You can tell what recipes my mother used by where the book falls open and where the stained pages are.
posted by Dip Flash at 5:37 PM on January 5


One point that remains consistent across virtually every nutrition and health recommendation is that eating home-cooked meals prepared with fresh ingredients correlates with better health.

....Which is a pretty convoluted way of saying that being rich correlates with better health.
posted by odinsdream at 5:39 PM on January 5 [4 favorites]


The edition I remember my mom having when I was growing up had whale recipes. Or, at least one. I can't remember.
posted by praemunire at 6:17 PM on January 5


there seems to be no reference to Goat?

IME you can prepare goat pretty similarly to lamb (or depending on the age of the animal, mutton, but it's pretty rare to come across mutton these days). It's good!
posted by kenko at 7:09 PM on January 5 [1 favorite]


We have one copy, rebound. Not sure of the date (maybe the 1951?), because it's in storage, with all of my other books.

Reading articles like this reinforces my dark sense that nutrition "science" is a shambles, and should be rebooted.
posted by doctornemo at 8:21 AM on January 6


Oh, we love the Gloria Bley Miller Chinese cookbook and have used it frequently, for years, kalessin .
(Speaking of rebinding, this is going to need some, as our copy has segmented into signatures or fascicles)
posted by doctornemo at 8:23 AM on January 6 [1 favorite]


I have one original printing, which like yours is falling apart. We have another, newer printing that we generally use for reference, not having had the older one rebound, but perhaps that would be a good project to pursue.
posted by kalessin at 10:53 AM on January 6


I'm not going to touch Wansink's rep as a whole, but I didn't see that this particular thing was debunked, re: Joy.

"He picked recipes that were just one cake and not servings!" but how were the calories calculated then? Did calories per cake go up? I'm interested.

Some other good points were made about the low percentage of recipes tallied, but I thought the complaint about him only calculating recipes with the same name was a bit facile. I guarantee if he hadn't, there would have been complaints about that too.

Clearly the guy is untrustworthy, but I'd like to actually know how the old Joys compare because I think it's interesting.

Skeptical personally, because I remember my grandmother's cooking as very very fatty indeed, but that's a sample of one.

Does anyone know where I can get my hands on older joy recipes without searching out and buying all the older cookbooks? I have one from the 70s, and I'm sure the library has the newer ones.
posted by liminal_shadows at 1:24 PM on January 6


I'm not going to touch Wansink's rep as a whole, but I didn't see that this particular thing was debunked, re: Joy.

"He picked recipes that were just one cake and not servings!" but how were the calories calculated then? Did calories per cake go up? I'm interested.


Eh. A full-size Snickers has more calories than a fun-size Snickers, but it would still be disingenuous to claim that "eating candy from the checkout lane is 300% worse for you than eating Halloween candy" on this basis.

Bottom line, his methods were faulty and he has an incredibly shady record. At this point the burden of proof should be on anyone seeking to back up his claims, not on those debunking them. No scientific claim that he has made should be treated as valid until proven so verifiably and independently.
posted by duffell at 1:39 PM on January 6 [2 favorites]


I do you one further. Not just Wansink's, but no one's scientific claim should be treated as valid until proven verifiably and independently. That's part of the scientific method: reproducibility.
posted by kalessin at 3:43 PM on January 6 [3 favorites]


Legit.
posted by duffell at 4:17 PM on January 6


« Older Wherever you go, there they are: A short history...   |   The Legend of Zelda has a Minus World Newer »


This thread has been archived and is closed to new comments