Morals keep changing: Words do not reflect trends

The research on changing morals by three Australian psychologists is an interesting attempt. It is good to see some robust claims about attitudes related to morality through empirical research as against abstract philosophical writings on morality.

But doing this kind of empirical research also has a drawback when it comes to making claims about morality across communities and societies since results arising from such studies also reflect limitations of methods used in the research.

The basic conclusion of this work is potentially interesting and perhaps counterintuitive: we are more compassionate than we were earlier as a society. The ‘we’ should be taken with a pinch of salt because although the authors might want to suggest that the ‘we’ includes all humans, their research is relevant at the most to the western world.

They begin by noting that “values such as compassion, care and safety are more important to us now” and “judging right and wrong based on loyalty has steadily risen”.

The authors were interested in how morality changed over time by tracking changes in “patterns of language use”. They used the rich source of Google books which has over 5 million books to analyse the change in language terms related to morality over time. Results from this analysis showed that from the beginning of the 20th century there was a strong shift to demoralisation, meaning less prudish and less judgmental terms were used in the books. But from the 1980s onwards, they found that ‘our culture’ has become more judgmental and censorious. They also note that the “decades since 1980 can be seen as a period when moral concerns experienced a revival”.

They speculate that this could be because of the election of conservative governments in the US, UK and Australia during this period as well as the importance of “political correctness” related to social justice concerns. The authors do recognise that this is based on data that is largely from American sources but they believe that such methods do indicate how trends in our moral thinking over time can be mapped.

While this attempt to map changing moral values of a society is interesting, there are some fundamental problems in assumptions that make such a study possible.

First of all, what does it mean to say that there is a morality associated with a group of people? To think that all people in Kerala hold some type of moral values is questionable. The fact that the authors speak as if a collective holds common moral positions has a silent assumption, namely, that moral values are something we universally hold. The belief that we hold some universal values of morality is one that has been influential in understanding morality.

However, we also know the pitfalls in this belief. It could be, and in my view more often, the case that our moral judgments depend on the context and not on some underlying value or principle. Even the basic moral value of not killing another is violated when it comes to war, death penalty, self-defence and so on. This belief in these enduring moral principles also indicates the cultural bias of the researchers. Many communities, including approaches from many Indian traditions including Buddhism and Jainism, look at moral judgments as being dependent on individual contexts. Thus, it is that the philosophers of these traditions do not bother to construct universal theories of morality but teach morality through hundreds of stories situated in contexts.

Another important problem is the method of coming to this conclusion. Tracking moral values through the use of moral terms in published books (those that are in the Google database) is interesting but can say little about social positions on morality.

First of all, these books do not represent society in any way. Given that writers, especially those privileged ones who actually get published, are somehow the voice of the larger society is an unreasonable belief. Many writers are from the margins of society or who are academics whose work many times is quite disjoint with the experience of the larger society.

Books, especially in earlier times, do represent a literary elite and even when they wrote about ‘other people’, it is questionable whether they were able to map the moral world of these people in their published work. We should also remember that these are published books and publishing is also socially skewed. Who gets published and what gets published reflects not opinions of society but only of a few business interests. So how can these books represent anything about the larger society?

The lack of published books by the marginalised and subaltern should also be taken into account in this exercise. As we know, moral terms in subaltern expressions may be very different from gentrified versions. And if we search only for patterns of some “mainstream” words, then we miss any real understanding of the moral world of the larger society.    

Moreover, uniformity in the use of certain moral terms may say little about holding equivalent moral beliefs. We tend to use terms because we have been taught to use them in certain contexts, and not because we understand their full meaning. Certain words get currency and get widely used because of their use in schools and colleges or in the media.

This is quite independent of the meaning of those terms! Moreover, there is another problem.

If I use the word “equality” many times in my writing, it does not mean that I subscribe to equality. I may actually be very critical of equality but when we search for words (without unpacking the context), it will appear as if my use of equality makes me a proponent of equality.

Frankly, I find it quite meaningless to correlate the amount of words used in a text and the belief systems of a society. Language does not work this way. There is a similar case in the Indian context when some writers argued that the vacanas — a special kind of literary compositions in the Kannada language — were not about caste because the word ‘caste’ was not explicitly used in any significant manner. Such arguments about finding social reality in the use of words in texts are a fatal flaw in understanding texts. For example, scientific texts rarely, if ever, use the word reality but to think that the discourse of science is not about the real would be a serious mistake since they are only about the nature of the real.

Finally, there is another fundamental problem related to language in this exercise. Patterns search for similar words used across time. So when they find the word ‘loyalty’ appearing more in a certain time period as compared to another period, they conclude that the question of loyalty was more important in the society in the former case. However, this argument is flawed because of the way meaning changes over time. Words do not have the same meaning over time.

There are countless examples of how the meaning of the same word radically changes over time. The meaning of terms such as ‘just’ or ‘equality’ is not the same as it was 10 or 20 years ago. Meaning accrues through its uses and contexts. So to conclude anything — especially something as important as moral beliefs of a society — from such exercises is problematic.

— (Sundar Sarukkai is a philosopher based in Bengaluru)