Semantic Inflation Can Create False Trends

The rise in anxiety can look downright depressing.
But maybe it’s an artifact of “concept creep” instead.

What do you mean by depression? That’s a question that needs to be asked by psychologists who report alarming rises of anxiety and depression over time.

Semantic Inflation: A Creepy Concept

Have the concepts of ‘anxiety’ and ‘depression’ been normalized or pathologized? A corpus study of historical semantic change (Xiao et al., PLoS One, 29 June 2023).

Four psychologists at the University of Melbourne, Australia, were curious about reported rises of anxiety and depression: are they real, or artifacts of semantics? They knew that this has happened with other psychological terms, in which shifting definitions create false trends. “Concept creep” is a creepy concept that can lead to poor science, where artifacts of word usage changes can look like crises.

Research on concept creep indicates that the meanings of some psychological concepts have broadened in recent decades. Some mental health-related concepts such as ‘trauma’, for example, have acquired more expansive meanings and come to refer to a wider range of events and experiences. ‘Anxiety’ and ‘depression’ may have undergone similar semantic inflation, driven by rising public attention and awareness. Critics have argued that everyday emotional experiences are increasingly pathologized, so that ‘depression’ and ‘anxiety’ have broadened to include sub-clinical experiences of sadness and worry.

Again, what do psychologists mean by depression? More people might be suffering from it, or perhaps nothing has changed except the definition. Who on earth has not had episodes of sadness and worry? If those “sub-clinical” cases get lumped into the statistics, then the whole world is in a pandemic of depression! Aaagh!

Not to worry. The four psychologists looked at the records and found evidence of concept creep. They searched through 133 million words from psychology papers between 1970 and 2018, and more than 500 million words from general texts from the USA. For a null hypothesis, they predicted a decline in severity of symptoms in the period. The results showed otherwise:

We hypothesized that collocates of ‘anxiety’ and ‘depression’ would decline in average emotional severity over the study period. Contrary to prediction, the average severity of collocates for both words increased in both corpora, possibly due to growing clinical framing of the two concepts. The study findings therefore do not support a historical decline in the severity of ‘anxiety’ and ‘depression’ but do provide evidence for a rise in their pathologization.

This is like saying that if you are sad to have lost at tic-tac-toe or a paper-scissors-rock game, you are suffering from pathological depression. If you are worried about how you did on a test, you need to see a psychologist for pathological anxiety. Good grief; this is not what was meant by grief decades ago! If everybody is depressed, then “clinical depression” has lost its meaning. “Semantic inflation” corresponds to “semantic dilution” just like in monetary inflation. A dollar doesn’t go as far in an inflating economy, and a concept doesn’t mean as much when too many words are chasing too few meanings.

Concept Creep Muddles Realism

To be sure, the researchers could not judge from their study whether so-called clinical depression and clinical anxiety were on the rise. “There may or may not be a relationship between the prevalence of a clinical phenomenon and the conditional probability of clinical terminology being used when it is mentioned,” they cautioned. They did not find evidence for semantic dilution in the literature for the two words. They did, however, notice an increase in pathologization: the tendency to categorize sub-clinical cases of sadness or worry as disease states. In their words, “the strong trend in both psychological and general discourse has been to place a clinical frame around them.” Ordinary sadness must not be “normal” in this instance of concept creep. You need to see a shrink!

A tendency to use pathological or diagnostic language to make sense of everyday distress might be one way in which vertical creep takes place. Even so, the trends we have identified are better described as pathologizing rather than as vertical concept creep (normalizing).

By vertical concept creep (normalizing), they mean changing the level of anxiety that is considered normal. What they found is the opposite: semantic inflation makes more people appear abnormal. Either way, shifting definitions cloud the possibility of doing good science.

Using very large text corpora representing academic psychology and general culture (USA blogs, fiction, magazines, newspapers, spoken language, TV), we found that the concepts of ‘anxiety’ and ‘depression’ have undergone notable shifts in their emotional meaning. Contrary to the hypothesis that they would become increasingly associated with less intense, severe, or harm-related words, the opposite pattern consistently emerged across concepts and corpora. That pattern appeared to reflect, in part, a rising tendency to pathologize ‘anxiety’ and ‘depression’ by locating them in the semantic context of diagnosis, disorder, and symptoms. The cultural implications of this trend, and how it relates to the broad pattern of concept creep, remains to be determined.

Creeping Concepts Call for Clarity

The specific instance discussed in the paper calls for more clarity in all scientific research. More generally, it relates to a long-standing issue in the philosophy of science concerning meaning and reference. The prophet Amos used a proverb to illustrate the point, “Do two walk together, unless they have agreed to meet?” (Amos 3:3). Two parties need to agree on what they mean by a scientific term in order to make sense of an argument (see Equivocation in the Baloney Detector). Philosopher of science Jeffrey Kasser said, “The greater we extend our semantic search, the more risk we run of exceeding our epistemic grasp.”

Lesson: when someone uses a term to argue for a position, a good question is to ask what they mean by the term. For instance, Evolution is a term fraught with many possible definitions: unfolding, elaboration, emission, programmed development, microevolution, or molecules-to-man common ancestry by natural selection (Darwinian evolution). “What do you mean by evolution?” Without prior agreement on the meaning of the word, two parties can talk past each other.

Exercise: Look for other scientific terms that can cause confusion. Find historical examples of concept creep or semantic inflation. For starters, consider words like gender, empirical, evidence, or explanation. Do scientists mean the same thing today that they meant in Victorian Britain with words such as reptile, avian, species, genetic, related, model, cause, determine, natural, or proven? What debates today boil down to parties talking past each other because of unclear definitions?