Category Archives: research

Holism, Reductionism, and Combos® Baked Snacks

Many, many years ago — when I was 23 — I saw a very compelling episode of the public television show Nova called “How to Create a Junk Food.” This was a documentary from Britain’s Channel 4 that was given a new voice-over by the producers of Nova, as well as a silly intro and outro featuring Julia Child. The show chronicled the development of a new snack food.

This past year I’ve watched this documentary twice, once in San Diego through the miracle of inter-library loan and once in the Media Resources Center in UC Berkeley’s Moffitt Library. The program tells a remarkable story of technology gone wrong. In the 1980’s, the British company PA Designs Ltd. hoped to fill a snack food market niche. Their market research showed that consumers wanted an easy-to-eat snack that was healthy, natural, and had no chemical preservatives. They wanted something that was nutritious, tasty, and crispy.

The food technologists at PA Designs started with this mandate, and through a careful, step-by-step evolutionary process — addressing various problems of manufacturing requirements, consumer preferences, and cost restrictions — they developed a brand new snack, one that eventually came to America as this:

Combos

What are those, you ask? Why, those are Combos® baked snacks.

Healthy? Natural? Nutritious? No chemical preservatives? What the heck happened?


Initially, the food scientists thought that they could make a crispy cone baked from croissant dough, and fill it with moist savory fillings. Peter Sampson, of PA Designs, immediately noted a potential problem when he and his crew of food technologists were sampling fillings — beef, and cheese, and chicken-and-prawns — that had been prepared in their test kitchens. He said,

At this stage, […] all of this looks so edible and so nice — but then, it looks familiar. The skill here, I think, is going to be in maintaining as much of that presentation as we can when it becomes a sludge that we extrude into that [outer shell].

As it happens, even the initial prototype versions of the snack made in the test kitchens did not go over well with consumers — the British housewives who sampled the snacks agreed that they were too messy, and were not suitable for eating without knife and fork.

To address this problem, the food technologists decided to move away from the baked croissant cone to an extruded product, perhaps made with a newly-developed cooker-extruder. And, because the flavors of the fillings did not stand up to the high temperatures needed for pasteurization, they decided to enhance the fillings with natural and artificial flavors — or, as they say in Britain, “nature equivalent” flavors.

Slowly, step by step, driven by the problems they faced, the food technologists moved ever further away from the “healthy, natural, nutritious” food that they had intended to produce, and created a Frankenstein’s monster.


What went wrong? It is the curse of the scientist, of the person with a reductionist world-view: taking care of the minutiae, without keeping an eye on the holistic big picture. What seems tragic to me here is that the ideas that the food technologists used are ones that I use myself all the time while cooking, ideas that I think are in fact necessary in order to cook well and to be comfortable in the kitchen.

You taste the sauce, and note that it is not sharp enough, so you think of acidic ingredients that you can add: lemon juice? vinegar? You want to make some hash with your kale, but you have no potatoes, so you think of substitute starches — leftover rice, or pasta… or maybe you quickly toast some bread and cut it into small cubes. Solving problems like an engineer is part what you do when you adapt recipes on the fly.

But with too many substitutions, too many improvisations, you might create a dish that is not recognizably connected to what you walked into the kitchen thinking of. It’s OK to think of the whole as being the sum of its parts; but little changes to each of the parts can add up to a big change for the whole.


The marketers at PA Designs came up with several names and packaging options for the extruded product the food engineers eventually produced. The high-end option was “Mousse en Croûte.” An option meant to appeal to British tradition was “Mrs. Palfrey’s Cracked Wheat Savouries.” The name they eventually settled on was “Crack-a-Snack.”

I don’t know whether Crack-a-Snacks were ever actually produced in England. But the concept of an extruded tube filled with a savory filling found its way to the United States in the form of Combos®; this television ad introduces them to an innocent nation, while suggesting that the cheesy filling is inserted by hand. Nowadays, when I see the packages of Combos® on display in the gas station mini mart, I always think to myself: Hmm, mousse en croûte.


(Here is some extra bonus material about the Nova documentary. First, a review in New Scientist of the Channel 4 documentary; second, a review from the Los Angeles Times of the Nova episode; and finally, perhaps only available until the lawyers from WGBH in Boston notice that it exists… the final 35 minutes of the Nova episode, ripped from a Betamax tape recorded in 1988 from a KQED San Francisco broadcast.)


Image of Combos® snacks taken by the author. Featured image: Detail from Frankenstein observing the first stirrings of his creature, engraving by W. Chevalier after Th. von Holst, 1831. More information here.

This Modern World

Every now and again, I am freshly amazed by what a person can access through the internet. I’ve become blasé about some things: finding obscure mathematical papers online, for example.1 But yesterday I was struck by how much easier it has become, even in just the past few years, to find information that used to require physically traveling to distant dusty archives.

For example…


In the fall of 2004 I was traveling in England for work, and I had a day or two in London. I stopped for fun at a shop that sells antiquarian books and prints,2 and I looked through their various collections. In a section devoted to botanical prints, I found a nice old print that did not cost too much and that I thought might look nice in our home. Here it is:

Buchoz

All I knew about it when I purchased it was that it was drawn by Pierre-Joseph Buc’hoz, that it was from the late 18th century, and that (as one can see from the upper left corner) it was plate number 87 from some collection.

After I got home I tried to find out more about this print. Using the power of the internet in 2004 — mainly, online catalogs from libraries — I figured that it was most likely from a book by Buc’hoz entitled Collection précieuse et enluminée des fleurs les plus belles et les plus curieuses qui se cultivent tant dans les jardins de la Chine que dans ceux de l’Europe, and published in the late 1770s. But the only copies of this book in the United States seemed to be in libraries on the East Coast, not so convenient for me to go to.

Within the following year, though, I happened to be in the Washington DC area for work… so one day I went off to the Library of Congress. I got permission to go to their special collection room, and a clerk fetched the two volumes of Buc’hoz’s Collection précieuse et enluminée des fleurs for my perusal. No cameras were allowed.

I checked: and yes! The print in our study was indeed plate number 87 from volume one of this book. Case closed. And by looking at the list of plate titles at the back of volume two, I found that our print was of “La Sparmann de la Chine”, as described in Buc’hoz’s Collection des Plantes nouvellement découvertes.3


Yesterday we were visited by someone who I thought might have a professional interest in this print by Buc’hoz, so I dug out the notes I had taken at the Library of Congress. And then it occurred to me that in the decade subsequent to my visit to Washington, perhaps someone had scanned some more of these prints.

A quick Google search for the title of Buc’hoz’s book led me to the Bibliothèque nationale de France (whose online collection I am quite familiar with, because they have scans of many 18th- and 19th-century mathematics books and journals). Lo and behold: high-quality scans of Buc’hoz’s book, both volume one and volume two!

Without leaving the living room couch, I could get information that, just ten years ago, had required visiting the Library of Congress. And the passage of time was indeed necessary to make this search easier: the Bibliothèque nationale only digitized these books last year.


I’ll close this post simply by noting: They really knew how to make a title page, back in the day…

Buc'hoz


  1. When I was editing this book I tried to find online versions of the papers cited in the bibliographies of each chapter, so that we could provide links to them in the pdf version of the book. I was particularly proud to have tracked down this online version of a paper from a defunct journal that almost no libraries have copies of; this digital copy was scanned from a physical copy owned by the New York Botanical Society. 
  2. I think it was Henry Sotheran Limited on Sackville Street. 
  3. “Nouvellement découvertes” by whom, one might ask of the French botanists.