Nathan Glazer here reports on a remarkable study of the frequency of insanity in the limited States, from 1840 to the present, that tends to shake views held by almost everyone.
The various social sciences, like all disciplines possessing an individual history and a corps of specially trained practitioners, ask their own questions, and answer them in their own way. It is not often that the questions they ask are the layman’s questions, or the answers they give ones that would satisfy a layman. When they ask a question which has served for decades as one of the common counters in discussions of modern life, and at the same time answer it—that is news.
It is in just this sense that a slim volume recently published by the Free Press (Psychosis and Civilization, by Herbert Goldhamer and Alexander Marshall, 126 pp., $4.00) is news. The question it asks is: is it true that the frequency of mental disorder—specifically, of psychoses—has increased over the past hundred years? We know that enormous changes have occurred in the way we live in these hundred years. Many more of us live in cities, the cities are larger and noisier, we travel greatter distances to and from work, are subject in larger measure to the tyranny of the clock and the need to oblige a superior—and in view of all this, it would appear a truism to assert that man, subjected to an increasingly inhuman (or at any rate nonhuman) environment, increasingly breaks down under the strain. And indeed, all around us are huge installations which we know house many thousands of the mentally disordered, and the budgets of state governments groan under the pressure of maintaining them and building more. Surely all this, if not new, is far more characteristic of our present-day lives than of life a hundred years ago. But are we sure?
This would seem to be a question on which we cannot achieve certainty. For it does not appear very likely that we could find statistics for the 1840’s and 1850’s which would enable us to answer such a question. Previous studies had, indeed, not gone back very far —hence it was no surprise to find that some of them had discovered no change in the frequency of mental disease. The one study that went furthest back (to 1881, in Massachusetts) did show a rather large increase.
Goldhamer and Marshall decided to go back even further, and in effect constructed their own statistics on the basis of data from the state of Massachusetts beginning with the year 1840. At that time, there were only three institutions in the state devoted specifically to the insane—a private hospital in Boston, established in 1818 (McLean), the Worcester Hospital, then only eight years old, and the South Boston Hospital, then only one year old. In addition, the insane were received in state and town almshouses and in the prisons.
In view of these limited facilities, could the data on the insane be very complete or reliable? Goldhamer and Marshall argue that they were, and very effectively. Quoting from the early reports of the asylums and their superintendents, they demonstrate, for example, that a very clear distinction was made between the psychotic (then called “lunatics”) and the mentally defective (then called “idiots”). The latter were almost entirely excluded from the limited places available in the asylums. In the same way, those suffering from epilepsy without psychosis and alcoholism without psychosis were also clearly demarcated from the psychotic, and for the most part excluded. The detailed descriptions of the disorders from which the lunatics of the 1840’s and 1850’s suffered indicate clearly that we deal with the very same diseases —even though “mania,” “melancholia,” “dementia,” and “monomania of suspicion” had to make do, in those days, to describe what we currently call “manic-depressive psychosis,” “schizophrenia,” and “paranoia.” The authors carefully exclude the possibilities of error related to the readmission of patients (all of the figures quoted here refer to first admissions), or to the chance that the mentally ill of Massachusetts went out of the state for treatment (not likely, in view of the fact that Massachusetts then—as now—had facilities for the insane as good or better than any other state in the union), or to the fact that the hospitals of Massachusetts contained out-of-state patients (they did, but the record-keeping was excellent, and these can be excluded).
In short, with a most exemplary and remarkable care and precision, the authors of this study establish rates for the frequency of admission for psychosis in Massachusetts for the years 1840 to 1885, rates that must be very close to the true ones and which in any case will not be easily improved upon.
The rates thus established rise from 41 per 100,000 for 1840-45 to 58 per 100,000 in 1880-84. (This rise is undoubtedly a product of the increase in facilities during the period.) These rates seem at first glance much lower than the rate of admissions in 1940: the 1940 admission rate, applied to a population with the age structure of that of the 1840’s, would give a rate today of 85 per 100,000, and applied to the population of the 1880’s, 91 per 100,000.
However, when one breaks down these overall rates of admission to examine the rate for each age group, a remarkable and most meaningful difference in the pattern of admissions appears. The 1940 rate rises rapidly for the ages from ten to about thirty, then is about the same for those aged from thirty to fifty, then begins to rise rapidly again, to become very high for the old. In the 19th century, we find the same rise in the rate of admissions for the young, the same plateau between the ages of thirty and fifty, and then a drop in the rate of admissions past the age of fifty. The differences between the 19th-century period and our own day are created entirely by the forge number of admissions for psychoses of those older than fifty today. Up to the age of fifty, the rates a hundred years ago and today are roughly the same. (As we shall see later, there is a technical reason having to do with change in admissions practices that accounts for the rise in the over-fifty group.)
It seems almost inconceivable that this should be the case. Even if there has been no real increase in the frequency of psychosis, one would expect a great increase in the number of hospitalized psychotics. For, comparing the situation today with that in even as late a year as 1885, we find that facilities are much more plentiful and easily available, that the popular attitude toward hospitalization for mental disease is much more favorable, that nonpsychotic disorders such as alcoholism, idiocy, and even psychoneuroses are increasingly dealt with in mental hospitals.
Yet, no matter what our bases of comparison, the frequency of psychosis in the middle of the 19th century is not less than it is today. The earliest rates on record for Massachusetts, for 1840-44, when, as we have indicated, there were few facilities and those only in existence a few years, exceed, for ages thirty to sixty, those for the state of Maine in 1940. And the admission rates for psychosis in Suffolk County—that is, Boston—were in 1875-79 higher than they were in New York City in 1929-31, for all ages up to fifty-five! Even as early as 1840-45, the Suffolk County rate was higher than the New York City rate for the age group forty to sixty. A comparison of the rates in 1885 and 1930 “reveals that the male 1885 rate for the . . . age group 20-40 slightly exceeds that of the contemporary period and that the 1885 female rates for ages 20-50 exceed the corresponding 1930 figures.”
Nor do we deal with an exceptional situation when we deal with Massachusetts. The authors carefully consider any factor that might have tended to lead to a higher frequency of mental disease in Massachusetts in the period from 1840 to 1885. Thus, they examine the bearing of the proportion of foreign-born, which was low at the beginning of this period (5 per cent in 1840) but rose rapidly, with the Irish immigration, to become quite high by the end of the period (1860, 21 per cent; 1870, 24 per cent; 1880, 25 per cent). However, this proportion of foreign-born has since been fairly constant (it was also 25 per cent in 1930) and could therefore not have been the cause of a peculiarly high rate of psychosis in the middle of the 19th century.
Nor is it the fact that Massachusetts was, among all the states of the union, particularly prone to mental disorder in the middle of the 19th century. Wherever we can find figures, we will find about as much psychosis—that is to say, admissions to hospitals for psychosis— in the 19th century as in the 20th. Thus, Oneida County, in New York State, had easy access to a state hospital established in Utica in 1843. The rate of admissions from 1843 to 1865 for the age group from thirty to fifty is only slightly below the 1930 New York State rate (adjusted for the rural-urban proportion that prevailed in Oneida County during this period). Perhaps most remarkable of all is the case of Fayette County, Kentucky, which had access to the hospital in Lexington, and for which we have figures from as early as 1824. For the years 1824 to 1842 (at the earlier date, Kentucky was scarcely out of the frontier stage), this county “had [a] higher total admission rate in the central age groups than . . . Kentucky . . . today.”
Even before we begin to try to understand why there should have been as much psychosis in the middle years of the 19th century as there is today, we have to consider how it was physically possible for admissions for psychosis to the few institutions that existed in the 1840’s to give a rate, even if only for certain age groups, equal to that of today.
There are a number of factors pointed to by Goldhamer and Marshall that are relevant. The first institutions opened specifically for the insane were established for those who created the worst problems for society—thus, the Worcester Hospital in Massachusetts was specifically limited to the “violent and furious.” Those psychoses associated with senility, which are responsible for such a large proportion of the inmates of present-day mental hospitals, were in effect ruled out by this definition, and it was only much later that such persons were taken into hospitals. Idiots and those of subnormal intelligence were also ruled out. In ef-
fect, these hospitals were established for persons afflicted with manic-depressive psychosis, paranoia, and some of the forms of schizophrenia. These are diseases which characteristically strike young people and people in the middle years of life. It was such people who filled the new hospitals established in the 30’s, 40’s, and 50’s of the 19th century. The psychoses of senility, which are responsible for the presence of most of the aged in our hospitals, and subnormal intelligence, which is responsible for a large proportion of the young, were both managed within the family. And since families were larger, and their living quarters also larger, there was no great tendency to hospitalize these unfortunates.
Consequently, Goldhamer and Marshall argue that even the great increase which has taken place in the last century in the proportions of old people hospitalized for psychosis is an increase only in hospitalization, not in actual frequency. Sufferers from the psychoses of senility, they argue, quoting contemporary records and the advice of leading doctors of the time, remained at home, and there was little pressure to get them into hospitals.
Similarly, the much smaller increase that has occurred in the numbers of young people (under twenty) hospitalized does not indicate a real increase of psychosis in this group, for a very large proportion of those in this age group today are hospitalized for non-psychotic conditions.
In other words: even where we see an apparent increase over a hundred years ago in the numbers in our state hospitals today (that is, among those under twenty and over-fifty), the increase can be explained by the expansion of facilities.
What are we to conclude from all this? The authors of this study are extremely cautious. All we can conclude, they say, is that there has been no great change in the conditions causing psychosis in this country in the past hundred years. Possibly if we could carry our study back another hundred years we would find a great change; possibly by 1840 those conditions of modern life that people believe lead to mental disorder had already been established. But it is not likely that any statistical studies can carry us back before 1840. It is also possible there has been a great increase in the frequency of neurosis; but this is a difficult matter to test statistically.
Another conclusion, they point out, is compatible with these results: that is, that in psychosis we deal with a condition which is independent of environmental circumstances, a condition dependent on heredity or physiological aberration, which, like some physical diseases, strikes a certain proportion of the population. The authors refer to a recent study of the frequency of psychosis on the Danish island of Bornholm, an island inhabited principally by farmers and fishermen. It turns out that the frequency of psychosis there is roughly what we find in New York State. Were it not for this finding, one might be tempted to suggest that the trauma of leaving one’s home and emigrating across the ocean had affected Americans in general so deeply that the same proportion of psychosis existed among frontiersmen as among city dwellers. However, the Bornholm study shows us that even where the population has deep roots, we find the same measure of mental disorder as in America.
We do not yet know enough to be able to do more than speculate as to the implications of these findings. We already know enough, however, to lay to rest one of the most popular clichés of our culture, one that we run into again and again in sociological, political, and religious writing, and which has often served as the basis for very lazy conclusions. We may hope that the Goldhamer and Marshall study will be only the first of a series analyzing the records of different countries, and different times.