Why People Smoke

Carl C. Seltzer has been engaged in research in physical anthropology at Harvard since 1937, and a survey which he made of the smoking habits of the class of 1946 led to some of the findings in this present article. At present, Dr. Seltzer is Research Fellow in Physical Anthropology at the Peabody Museum and Research Associate in the Adolescent Division of the Children's Hospital

Whether or not there are smoker personality types, there may be important underlying processes which are in part responsible for the differences between smokers and nonsmokers. It is conceivable that such processes may play a role in determining whether a person becomes a smoker, and even the form of smoking adopted.

The eminent biometrician, Sir Ronald Fisher, reported on a study of the smoking habits of a series of male twins based on data gathered in Germany. His data showed about twice as many identical twins to be alike in their smoking habits as nonidentical twins—65 percent against 33 percent. In another group, of female twins from England, 83 percent of the identical pairs were alike in their smoking habits, as compared with 50 percent of the nonidentical pairs.

Further analysis of the female-twin data, in order to eliminate the possible mutual influence between twins living together, revealed greater concordance of smoking habits in identical twins brought up separately from shortly after birth than in nonidentical twins. Similar results were also obtained, independently, from data gathered in Scandinavia. This evidence suggests that there is a genetic factor in the formation of patterns of smoking behavior. In a study just completed but as yet unpublished, I investigated the relationship between characteristics of physical structure and the smoking habits of the members of the Harvard class of 1946; Certain aspects of this study are noteworthy. The class of college undergraduates examined anthropometrically was surveyed sixteen years later for its past and present smoking histories. The availability of anthropometric data at the college age gave the special advantage of representing the "pristine" physical status of the individuals, unaffected by habit, diet, physical activity, health, and disease of the subsequent adult years.

Moreover, the period covered by their smoking histories was closely related to the age span of maximum smoking experience. At the time of their reply to the smoking questionnaire, the Harvard alumni were thirteen years out of college and averaged thirty-five years of age. And the number of individuals composing the study—more than 900—was large enough to permit the breakdown of the smoker series into exclusive groupings of "pure" cigarette, "pure" cigar, and "pure" pipe smokers. All "mixed" smokers—those who regularly used more than one form of tobacco—were omitted from these classifications. Thus, many of the criticisms leveled against other studies were obviated in this investigation.

Briefly, this study revealed that substantial differences in body build exist between smokers and nonsmokers. Smokers are consistently larger than the nonsmokers. They are taller, heavier, broader in the shoulders and hips, bigger in the size of the chest, leg, and hand. But even more significant, there is a consistent graded arrangement of physical differentiation among the smoker types. The pure cigarette smokers are the least differentiated from the nonsmokers in physical structure, followed by the pure pipe smokers, while the pure cigar smokers differ most from the nonsmoker group. For example, in the case of body weight, the pure cigarette smokers are on the average more than four pounds heavier than the nonsmokers, the pure pipe smokers more than six pounds heavier, with the pure cigar smokers averaging an amazing ten pounds more in body weight than the abstainers. Similar patterns are observed for a number of other body measurements, all indicative of the same trends of deviation.

Despite the inherent limitations in the research done thus far, it has become increasingly clear that smoking, the form of smoking adopted, and abstention from smoking are structured reflections of very complex forces, innate and environmental, in constant counterplay. Rather than a superficial habit overlaid indiscriminately upon various persons, smoking appears to be a response to a wide variety of personality and behavioral characteristics which have their origin, in part, in the biological or genetic makeup of the individual.

This opens up important and far-reaching implications for the highly controversial subject of the association of smoking and lung cancer.

If smoking has a constitutional basis, then there is a possibility that persons of a certain makeup are peculiarly liable to both smoking and lung cancer, as well as other diseases. This is not as large a speculative step as appears on the surface. That there is a relationship between constitution and disease is generally accepted in medical circles. Such a relationship has been indicated for an extensive variety of diseases, including coronary artery disease, diabetes, arthritis, duodenal ulcer, rheumatic fever, tuberculosis, and hypertension. Even with respect to cancer, the evidence favors a biological substrate for cancer of the breast, cervix, and uterus, and it is not beyond the realm of possibility that a similar situation pertains to cancer of the lung.

There is as yet no published evidence to this effect, but the possibility cannot be dismissed lightly. Further investigation remains to be done, and a comparative study of the constitutional aspects of lung-cancer patients will shed some light on the problem, one way or another.

An analogous supposition may have an even stronger basis in the case of the nonsmoker minority. Nonsmokers may be of a constitutional type that is, generally speaking, biologically disposed to rigid, inhibiting, self-protective habits, correlated with constitutional forces which resist disease.

This is not to imply that a determination of a constitutional lung-cancer factor will automatically solve the problem of the relationship of smoking to disease. Not at all. For it is highly unlikely by itself to account for the whole of the tenfold excess of lung-cancer mortality for cigarette smokers as compared with nonsmokers. In no sense would it overthrow the known statistical association between heavy cigarette smoking and cancer mortality, but it might serve to moderate the role of smoking as a lung-cancer risk. It would, moreover, establish the fact that the individual's constitution is an element to be reckoned with in the predisposition to lung cancer. We do not need to be persuaded of the concomitant rise in the mortality ratios for lung cancer and other diseases with increased amounts of cigarette consumption. This is a verified and accepted finding. No one is ruling out cigarette smoking as a prime suspect in the causation of lung cancer. But, for the present, we do have to consider the existence of accessories to the fact as well as possible accomplices. Pending the complete evidence, the problem of smoking and disease is still not fully settled.

Jump to comments
Presented by
Get Today's Top Stories in Your Inbox (preview)

What Is a City?

Cities are like nothing else on Earth.

Join the Discussion

After you comment, click Post. If you’re not already logged in you will be asked to log in or register. blog comments powered by Disqus


What Is a City?

Cities are like nothing else on Earth.


CrossFit Versus Yoga: Choose a Side

How a workout becomes a social identity


Is Technology Making Us Better Storytellers?

The minds behind House of Cards and The Moth weigh in.


A Short Film That Skewers Hollywood

A studio executive concocts an animated blockbuster. Who cares about the story?


In Online Dating, Everyone's a Little Bit Racist

The co-founder of OKCupid shares findings from his analysis of millions of users' data.


What Is a Sandwich?

We're overthinking sandwiches, so you don't have to.


Let's Talk About Not Smoking

Why does smoking maintain its allure? James Hamblin seeks the wisdom of a cool person.



More in National

More back issues, Sept 1995 to present.

Just In