Science, Society, and Kṛṣṇa Consciousness

By Krishna-kirti das, 25 January 2014

For many devotees, and especially for non-devotees, a doubt may arise as to the authority of the Vedic scriptures over the body of scientific evidence produced by the social sciences. Over the years, the hard sciences have produced many discoveries—powered flight, heart transplants, telephones, cars, computers, and so forth. In previous eras these would have been considered miraculous. For the social sciences, the expectation is held that if science is applied to the study of humanity similar miracles will be produced.

Some therefore feel that psychologists and sociologists could make discoveries that can be utilized in the service of Kṛṣṇa, much as how devotees use cars and phones. Problems like how to build a nuclear weapon, though difficult, are straightforward to solve, but problems like poverty or child abuse remain intractable. They always remain out of reach of a clear solution because the means of adequately understanding them solely by scientific observation and inference do not yet exist. Indeed, fields like sociology and psychology have yet to produce any kind of scientific breakthrough of the magnitude of those in the hard sciences. There is no sociological discovery equivalent to the splitting of the atom. The persistent lack of progress in the social sciences shows that human behavior is considerably more inaccessible to investigation than is inert matter.

This is where śāstra-pramāṇa matters. It can inform us of important dimensions of human nature that can never be discovered by empirical methods. And despite the best intentions, those who end up as professionals in the social sciences often promote or follow speculative theories that cause more harm than good. Research and social action guided by śāstra and bona fide representatives of the authorized guru-paramparā, however, does not have this problem.ⁱ

Causality and Bias in Science

Science is about understanding causality. An effect is studied to discover its cause. For example, before Louis Pasteur (1822 – 1895) there were popular theories about what caused infectious diseases, but none were useful in preventing them. However, Pasteur’s scientific demonstration of the germ theory of disease revolutionized the medical field. Among many important uses, this theory’s application allowed for greater prevention and control of epidemics. Pasteurization, the method of heat-treating liquids to prevent them from spoiling, is perhaps the most well-known of his germ theory-related inventions. Pasteur correctly identified and demonstrated a cause for some effect, and as a consequence an effective means of controlling disease was discovered. If a cause can be identified, its effect can be utilized, manipulated, or prevented.ⁱⁱ

Science applied to human affairs through the social sciences also seeks to discover causality. “What causes people to commit crimes?” “What causes marriages to break?” “What causes depression?” These are some of the many questions researched in this field. Finding causes for these problems ostensibly will make people happier, because the discovery of a cause will automatically suggest solutions. All public policies, therapies, and treatment plans that come from the social sciences depend on some idea of the cause for a problem that needs to be treated. Whether in the natural sciences or the social sciences, the goal of all scientific research is to understand causality.

Correlation and Causation

In order to discover causal relationships, it is necessary to remove bias from one’s research. Bias means that there is some known or unknown influence on an experiment that affects the results, and researchers do their best to control the influences known to them. However, the field of science not only requires that known biases be accounted for and controlled, it requires that unknown biases also be accounted for and controlled as well. Otherwise, test results cannot conclusively establish a causal relationship between one thing and another.

Causation means that something directly influences something else and not some other unknown cause. If an experiment is properly designed, it is the random assignment of test subjects to different conditions that assures that it is the behavior being studied and not some other, unknown cause.ⁱⁱⁱ Properly designed studies that randomly assign test subjects to treatments are the only kind of test that can demonstrate causality.

Otherwise, without randomization the most that can be achieved in a scientific study is the identification of a correlation between one thing and another. Correlation, sometimes called association, means that two or more characteristics are associated with each other, but their relationship is not necessarily causal.

It is often the case that studies which do not use randomization are misinterpreted as inferring a causal relationship when in fact they do not. This oversight is not uncommon, and it leads to erroneous conclusions about causality—even if there are many such studies and even if a firm consensus exists amongst them. This example from scientific literature illustrates the difference between correlation and causation as well as the ease with which a wrong idea about causality can be inferred from non-randomized studies:

“In a widely studied example, numerous epidemiological studies showed that women who were taking combined hormone replacement therapy (HRT) also had a lower-than-average incidence of coronary heart disease (CHD), leading doctors to propose that HRT was protective against CHD. But randomized controlled trials showed that HRT caused a small but statistically significant increase in risk of CHD. Re-analysis of the data from the epidemiological studies showed that women undertaking HRT were more likely to be from higher socio-economic groups (ABC1), with better-than-average diet and exercise regimens. The use of HRT and decreased incidence of coronary heart disease were coincident effects of a common cause (i.e. the benefits associated with a higher socioeconomic status), rather than cause and effect, as had been supposed.”^iv

The randomized trials demonstrated that there was some hidden variable (unknown factor) the researchers conducting the epidemiological studies had not considered. Once they reexamined their data in light of this discovery, they were able to identify important missing factors and bring their results in line with those of the randomized clinical trials.

Observational Studies

Scientific studies that do not use randomization are unable to establish causality. Such studies, including epidemiological studies, are called observational studies. Modifying the test subjects and being able to randomly assign them to different conditions is outside of the researchers’ control. Observational studies can identify correlations (which can be useful in suggesting areas of further research), but they cannot identify causation because they cannot rule out the possibility of influences unknown to researchers.

Sometimes observational studies have successfully demonstrated causality, but this has been rare. The studies that positively linked smoking to lung cancer provide a good example. It would have been unethical to have randomly assigned some people to chain smoke for years on end and others not in order to see which ones developed cancer. Some experiments did just that with animals. But animals are not people, so even those test results were not counted as sufficient to establish a causal link between smoking and cancer in humans. Thus observational studies were about the only kind of study that could be conducted.

The ethical concerns for testing smoking in humans limited the kinds of tests that could be conducted. That is why the effort to confirm the link between smoking and lung cancer took over 40 years and thousands of observational studies done on every dimension conceivably related to smoking. Because observational studies had to be used, researchers had the onerous task of demonstrating that no other, possible hidden variable could account for lung cancer except for smoking. Therefore a tremendous number of observational studies were conducted to account for every imaginable hidden variable. The magnitude of such an effort is exceptional, which is why it is found that observational studies are rarely used to establish causality.

Nonetheless, although observational studies are less reliable than randomized controlled studies, they are frequently undertaken because it would otherwise be impossible to conduct any other kind of study. Consequently, fields like the social sciences are almost exclusively limited to observational studies. Within the social sciences, the possibility of establishing causality with sufficient certainty by means of scientific observation and inference^v generally remains well out of reach.

Validity of Causal Theories in the Social Sciences

Because the social sciences are unable to produce within their problem domain a scientifically plausible demonstration of causality, researchers in these fields have necessarily had to borrow their notions of causality from the realm of philosophy, not science.

Some researchers have admitted that their key presuppositions about human consciousness and behavior are subjective, not scientific. The mid-20^th century pioneers of what is today known as “client-centered therapy” (Carl Rogers, Rollo May, Abram Maslow, etc.) were unafraid to argue that “an a priori understanding of human nature, whether consciously stated or not, was essential in the making of any psychology.”^vi Any therapy ultimately depends on notions of human nature that belong in part if not in full to the realm of philosophy, not science.^vii

In this particular approach to psychology, which remains popular today, the influence of existentialist philosophy is difficult to understate. Carl Rogers “thought that Kierkegaard’s insights and convictions expressed views he himself had held but was unable to formulate.” They had a “loosening up effect” on him. In the book Sickness unto Death, Kierkegaard argues that the aim of life is “to be that self which one truly is,” and Rogers understood this to mean that “one ought to allow one’s innermost nature to surface.” This idea was a cornerstone of Rogers’s thought on the self and on therapy.^viii

Aside from the field of psychology, sociology has an even more colorful background in speculative thought. In his seminal work, The Sociological Imagination, renowned sociologist C. Wright Mills writes that “so very much of modern social science has been a frequently unacknowledged debate with the work of Marx, and a reflection as well of the challenge of socialist movements and communist parties.”^ix

One characteristic of such thought is fascination with equality. Socialists of all stripes believe in what is called “class conflict,” which is the idea that exploitation of others is caused by inequality. According to this theory, people with more power than others will naturally try to protect their privileges, and in so doing, they will exploit those under their control. Therefore when sociologists research some subject that involves human suffering, they often look for inequalities and “differences in power” between one class and another. They typically recommend that some dubious form of “equality” be introduced in society to alleviate the problem.

The heavy reliance of the social sciences on philosophy for their causal theories shows that the studies attached to their theories, which are mainly observational, only give their theories the appearance of validity. This means that causal explanations from the śāstras have no less validity than speculative causal theories from the social sciences, which lack actual scientific confirmation. Popular speculative theories about human behavior promoted as if they were scientific are instead a reflection of some group of influential people’s unsubstantiated opinions, their “religious beliefs.”

The question then is which “religious beliefs”—doctrines—about human nature should be followed? The answer is that they must be authorized. Śrīla Baladeva Vidyābhuṣana in his Govinda-bhāṣya commentary on the Vedānta-sūtras discusses this:

“When we refer to a particular scripture, it must be authorized, and for this authority it must strictly follow the Vedic injunctions. If someone presents an alternative doctrine he himself has manufactured, that doctrine will prove itself useless, for any doctrine that tries to prove that Vedic evidence is meaningless immediately proves itself meaningless. The followers of the Vedas unanimously accept the authority of Manu and Parāśara in the disciplic succession. . . .”^x

Solutions based on varṇāśrama-dharma will be superior to whatever is currently on offer from the modern humanistic sciences. The application of scientific investigation to human affairs is welcomed, but it must also be guided by the śāstra-vidhi. It should not be speculative.

End Notes

i^ Fields like psychology and ethics never developed in India because they were obviated by the pervasive sense of dharma characteristic of traditional Indian society. Dharma subsumes these and much more.

ii^ This was only in the 19th century. People in India have been drinking hot milk for millennia, if only because that is what has always been done. Many dharma-śāstra authors also commented on, or wrote their own Ayurvedic treatises. The striking consistency between medical literature and dharma-śāstra is so well known, that S.G. Moghe can say (pgs. 30-31): “It will be proper to conclude here that in some respects, Dharma-Sastra, Ayurveda and Niti-Sastra are inseparably connected with each other.” (”Relation of Indian Medical Science [Ayurveda] to Dharma-Sastra,” in Studies in the Dharmasastra, New Delhi: Ajanta, 1991.)

iii^ Typical clinical drug trials randomly select participants into two groups, one with the drug being tested and the other without (the placebo group), because proof is needed that the new drug actually makes people better and not some other random or unaccounted for effect.

iv^ Lawlor DA, Davey Smith G, Ebrahim S (June 2004). “Commentary: the hormone replacement-coronary heart disease conundrum: is this the death of observational epidemiology?”. Int J Epidemiol 33 (3): 464–7. doi:10.1093/ije/dyh124. PMID 15166201. Qtd. in “Correlation does not imply causation.” Wikipedia. 3 Sep. 2013 <http://en.wikipedia.org/wiki/Correlation_does_not_imply_causation>.

v^ Pratyakṣa and anumān refer respectively to direct perception and logical inference.

vi^ Roy José DeCarvalho, The Founders of Humanistic Psychology (New York: Praeger Publishers, 1991) Questia, 21 Aug. 2013 < http://www.questia.com/PM.qst?a=o&d=14229715>.

vii^ “The keynote in the revolt and establishment of humanistic psychology was the understanding of human nature. The view of the person as a being in the process of becoming permeated the founders’ critique of behaviorism and psychoanalysis and dictated their views on method and psychotherapy. An a priori understanding of human nature, whether consciously stated or not, was essential in the making of any psychology, they argued. For this reason most psychologists of the time, especially behaviorists, regarded humanistic psychology as a philosophy or poetic psychology” (DeCarvalho).

viii^ DeCarvalho 62.

ix^ C. Wright Mills, The Sociological Imagination (Oxford University Press: Oxford, 1959) p. 82.

x^ Qtd. in CC Adi-lila 6.14-15 purport.