By Paul Rozin, published on November 1, 2000 - last reviewed on June 9, 2016
Paul Rozin Ph.D., looks at the psychologyof eating, culture and
The French have their cake and eat it, but don't show it their
bodies. Americans supposedly deny themselves the cake -- but show it on
their bodies nonetheless. This is the French paradox, the apparent
conflict between low cardiovascular disease in France and what we take to
be the rather unhealthful French diet.
The American dilemma actually has two parts: First, Americans think
the French situation is a paradox. Second, Americans are more concerned
about health and heart disease than the French (and try to do more about
them), and yet our rate of heart disease is about 30% higher.
On the face of it, this seems like a violation of Newton's second
law of mechanics (which deals with the conservation of energy)-something
that has more certainty and status than any claim ever made by a
But the French paradox isn't really a paradox at all.
In many ways, the French do eat more healthful diets than
Americans. Most critically, they eat fewer calories and a more varied
overall diet. Yes, they eat more grams of fat by and large, but the major
effect of fat is its contribution to excess calories.
Americans are the ones who have the problem with excess calories.
And even if the French ate an unhealthful diet, there would still be no
paradox. After all, the Dutch and Swedes eat what we would call an
unhealthful diet, and they live longer than Americans. The reason is
simple: Although there are relationships between diet and health, there
are many other determinants of health.
Partly persuaded by a sizable segment of the American medical
establishment, Americans believe that diet is the principal cause of
disease. But the biggest predictor of mortality is simply age. Diet
indeed contributes to health -- but so do genetics, lifestyle, quality of
medical care, quality of environment, stress, and many other
Compare our attitudes toward medicine, for example. The standard
American prescription for illness is antibiotics, diets and the like. For
the French, illness is seen primarily as an internal imbalance, and is
likely to be treated with relaxation, vacations and spa visits.
When it comes to diet, the fact is that we know very little about
why people eat as much as they do, or what makes them choose one food
over another. This is partly because psychology, as a field, has had very
little interest in food choice. We have focused instead on the amount of
food eaten -- in an attempt to understand that very visible condition,
obesity, which causes grief and compromises health. I think that as a
field, psychology has erred in both ignoring food choice, and in studying
food intake in nonoptimal ways.
Generally, researchers in this field (and I have been one of them,
to a modest extent) illustrate a very understandable predisposition in
scientists. We like clarity, simplicity and control. It is like the story
of the man who is looking for something at night under a street light.
Someone asks him what he is doing, and he says he is looking for his
keys. He is asked where he thinks he lost them, and he points, across the
street, to a much darker area. When asked why he doesn't look there, he
says because the light is here.
We have looked for the secrets of how meals begin and end, where we
have good control and techniques: we look at the physiological events
that precede and follow the meal. But I hold, as do a number of other
major investigators in this field, particularly Peter Herman, Ph.D., at
the University of Toronto, that the major determinants of how much we eat
during a meal are not physiological. And, by the way, they are not that
hard to study.
Perhaps the biggest determinant of how much we eat in a meal is how
much we are served. If food is reasonably palatable, we tend to eat what
is put in front of us. And here is part of the resolution of the French
paradox: French portion sizes are notably smaller than American portions.
This is particularly striking when comparing the size of ice cream
cones -- the French serve small golf ball-sized spheres while Americans get
tennis ball-sized mounds. The standard size individual portion of yogurt
in France is 125 grams; the standard size in America: 225 grams. Yes, the
French yogurt has somewhat more fat, but it has many fewer
Another major determinant of how much we eat during a meal is how
much we like the food. This is so obvious that it needs no
But another main determinant is less obvious. It has to do with
culture and situation. We eat at certain times and in certain contexts,
and typically eat a culturally prescribed meal. For Americans, lunch is
most likely a sandwich, a beverage, perhaps some chips and a dessert. We
eat it in the early afternoon, and we stop when we finish the dessert. Do
we stop primarily because we are not hungry? That's doubtful, since we
probably lost our hunger before we started the dessert. We're just
accustomed to ending meals with dessert. The French snack much less than
Americans, and do almost all of their eating at extended lunches and
dinners. When we are served an appropriate meal at an appropriate time,
we eat it.
My colleagues Sara Dow, Morris Moscovitch, and Suparna Rajaram and
I recently demonstrated the importance of this practice. We offered a
full lunch to two severely amnesic people, who have no memory of having
just eaten. After they completed lunch we removed the tray, and 10 or so
minutes later announced that lunch had arrived (a meal identical to the
first). On three occasions, the amnesiacs readily downed the second
serving. And both started to eat eagerly again when we brought a third
meal more than 10 minutes after they finished the second. When we tried
this with people with a normal memory, they looked at us as if we were
crazy when the second meal arrived, and said something like, "I just
ate." (Note: They didn't say they weren't hungry!).
There are other nonphysiological factors determining how much we
eat. These include the social context of eating, our beliefs about eating
and health, whether we are dieting, and so on. The point is that we can
spend another 30 years measuring blood glucose at the onset and
termination of meals, but we will find out only a small part of what
determines the size of a human meal. Indeed, several important metabolic
factors influence our eating, but they only modulate our meal size and
seem to work in a subtle way from meal to meal, day to day, and week to
When it comes to why we eat what we do -- namely, food choice -- the
situation is different. There has been very little work done on this
problem, perhaps a few percent of all of the studies on eating. It seems
obvious that culture is a major determinant of what we eat; does anyone
doubt that a Chinese person raised in an American household will have
American food habits?
Furthermore, there is not a simple "coin of the realm" for food
choice. Calories fill the bill for food intake. But for food choice,
there are also macronutrients (carbohydrates, fat and proteins) and a
variety of minerals and vitamins, all of which have to be consumed in
minimal amounts. The complexity of food choice and the clear involvement
of culture probably put psychologists off. But we cannot continue to
ignore the study of why we eat what we do. Food is a major part of our
life, a great source of pleasure, and in the U.S. in recent decades, a
major source of worry and stress.
There is no doubt that our culture has shaped how we view food. For
Americans and many others in the industrialized world, food intake has
undergone a major change during this century, giving rise to the
"American dilemma." There are at least five aspects of this
First, there is the epidemiological revolution. At the beginning of
the century, life expectancy was about half of what it is now, and most
deaths resulted from infectious diseases such as cholera, malaria,
influenza and the like. With the conquest of most of these diseases in
the industrialized world, along with public health measures, the major
health threats today are degenerative diseases such as coronary heart
disease, cancer, stroke and Alzheimer's disease. In the era of infectious
diseases, the delay between behaviors and illness was generally short.
But the behaviors related to degenerative diseases are often separated by
decades, making the consequences of our actions seem too distant to pose
a salient threat in the here and now.
Second, there is an excess of food. We were built to eat food when
it became available. We are now faced with a surfeit, in both amount and
variety, though we still don't embrace variety as the French do.
Third, there is the reduction in activity. Modern technology has
reduced the amount of physical effort that most people need to earn a
living. Of course, it's also provided a wide variety of labor-saving
devices, as well as forms of entertainment that require no physical
effort. So we're eating more and moving less -- a sure recipe for weight
Fourth, while there is an abundance of risk information linking
food and other habits to long-delayed positive or negative health
consequences, these relationships are often very small. They would never
be discoverable within an individual lifetime, so they become
Fifth, we are not educated to understand risks, benefits, nutrition
or the nature of science, all of which are critical for acting prudently
in the face of all the risk information we are exposed to. For example,
colleagues and I have discovered that a substantial minority of Americans
think of fat and salt as toxins, such that any amount of these necessary
nutrients is considered unhealthful. But, of course, that is
With the great publicity for diet-health links, and a great desire,
especially among women, to be thin, eating has become an ambivalent act.
True, it is necessary for life. True, it is often a source of great
pleasure. But, especially in America, it is also viewed as unhealthful,
in many ways. Almost everyone believes that not eating is riskier than
eating; but eating the "right" things and staying off the usually highly
palatable "wrong" things has become a daily stress in life.
Many have pointed to the prevalence of eating disorders, the fact
that obesity is actually increasing in the United States, and the fact
that our stores are simultaneously filled with "low cal" and "low fat"
foods, adjacent to tempting high-calorie, high-fat foods. The American
supermarket's yogurt section (unlike that of the French) is populated
almost entirely by low- or nonfat yogurt. But have a look at our
bakeries. The ambivalence is present in the store itself.
The evidence of ambivalence, or worse, about eating among American
women is abundant. In a questionnaire given to American college students
on six different campuses nationwide, Rebecca Bauer, Dana Catanese, and I
found that an average of 71% of females thought their thighs were too
fat, and 13% admitted that they would be embarrassed to buy (that is
correct: buy) a chocolate bar at a store. More so, ironically, than to
buy a condom!
Stimulated by all these differences, the "French paradox" and the
"American dilemma," my colleagues Claude Fischler in France and Sumio
Imada in Japan, and my students Allison Sarubin and Amy Wrzesniewski, set
out to measure the differences in attitudes toward eating, diet and
health in four countries: the United States, France, Flemish Belgium and
In each country we sampled both college students and adults waiting
in airports and train stations. We asked about the importance of food in
life, beliefs about diet and health, food choices, and modes of thinking
about food. I will focus on our analysis of the French-American contrast.
It suggests several factors that constitute the culturally distinct
relationships to food.
We gave subjects a choice of a one week vacation of luxury
accommodations with average food or average accommodations with excellent
food, both at the same price. Americans choose the better food
alternative almost half as often (42%) as the French (86%). We find that
in general, food brings to mind, in the French, the experience of eating,
while for Americans, it is more likely to conjure thoughts of calories,
nutrients, or the effects of food on the body.
It is particularly ironic that although the Americans do much more
worrying about food and health, and consume a much higher proportion of
foods that have been modified to reduce fat, a substantially larger
percent of the French (74%) see themselves as healthy eaters than do the
In all four cultures studied, women have a more negative attitude
toward food than men. It can not simply be attributed to greater concern
among women about weight and appearance; after all, we also see a more
negative attitude among women regarding factors affecting the food-health
link as well as the food-appearance link. The most extreme contrast
exists between the American female and the French male: she is troubled
by food, while he's content.
What we are seeing with these French-American differences about
food is probably something more general in the French-American contrast.
There is a tendency toward moderation in French culture, in contrast to
the American inclination toward excess.
Americans have a particular predisposition to spend a lot of money
on making their lives easier, and minimizing exercise or effort:
microwaves, air conditioners, power windows, automatic garage door
openers, driving to a store only a few blocks away. These are
expenditures that the economist Tibor Scitovsky calls "comforts."
The French spend much less money on such things, and as a result
get more exercise. They are more inclined to spend money on what
Scitovsky calls "pleasures": unique experiences such as fine meals,
plays, flowers, and conversation with friends. Scitovsky notes that
pleasures contribute more to happiness than do comforts; we appreciate
our air conditioning only when it breaks!
It's clear that French-American differences in milieu -- such as
reliance on cars and availability of snacks have a lot to do with
French-American differences in attitudes toward life and food. I doubt
that the French hypothalamus -- the part of the brain involved in food
intake -- is any different from that of the American. It's more a matter of
cultural values and styles of life. For the French, yum means only
moderate amounts of pleasure -- but for Americans the same word means piles
of food, and pleasure mixed with worry.
VIVE LA DIFFERENCE! A comparison of French-American attitudes
Legend for Chart:
B - FRENCH female
C - USA female
D - FRENCH male
E - USA male
B C D E
THE MODIFICATION OF DIET TO IMPROVE HEALTH
Frequency of eating low fat foods (% at least a few
times a week)
36 81 22 60
THE HEALTH VALUE OF FOOD
I rarely think about the long term effects of my diet
on health. (% true)
44 28 57 26
THE EXPERIENCE OF EATING
Heavy cream goes best with the word: whipped or unhealthy
(% who said whipped)
88 46 74 60
THE PLEASURE DERIVED FROM EATING
Enjoying food is one of the most important pleasure
in my life. (% true)
73 42 77 42
Paul Rozin, Ph.D., is a psychology professor at the University of