The Mismeasure of Intercultural Competence
By Milton J. Bennett
In his excellent book, The Mismeasure of Man, biologist and science historian Stephen Jay Gould points out the fallacy of assuming that “intelligence” is a measurable entity. I was struck that the same kind fallacy underlies popular approaches to measuring “intercultural competence.”
By tracing the history of IQ, Gould shows how observations become things – the process of reification. In the case of IQ, early 20th century French educators were trying to select children for different kinds of educational programs. They observed that some children were more inclined than others to engage schoolwork or solve certain kinds of problems. When they analyzed various measurements of that tendency, they named the constellation of scales that were most correlated with the abilities the “general intelligence factor” (g factor). Initially, g was just an indicator of the abilities, but then the measurement category was reified: it was assumed that “intelligence” was a thing that people had more or less of. IQ was then devised as a system to rank people in terms of how much g they had, and it is still with us today.
The point here is that g (and consequently IQ) has no identifiable existence outside of our measurement. We are, to state it strongly, inventing a thing to account for a consistency in our measurement. If you are following this, you might be asking, “so isn’t that true of all personal qualities and characteristics,” and you’d be asking a really good question that goes beyond the scope of this particular blog. Feel free to share your own speculation on that.
Here I just want to compare IQ to IC (intercultural competence). There is an eerie parallel. For instance, the idea of IC certainly began as an observation of behavior – some people seemed to be better than others at adapting to different cultural contexts. With this observation, two avenues opened up. I will get to the road less traveled later. The most popular avenue was the road to reification – assuming that people who were more competent had some set of measurable inherent qualities and characteristics that accounted for their competence. Once that assumption is made, the race is on to find or create scales that will “indicate” the assumed characteristic, and it is only a short step from there to assuming that the measured characteristic is, in fact, the competence itself. In this way, we have created a “c factor” that is to intercultural competence what the g factor is to intelligence.
Some people are even beginning to use the terms “cultural IQ” to refer to how much of their rendition of c people have. At least one measurement goes to far as to assume that IC is normally distributed in the population – just like IQ! This sets up a terrible contradiction. If intercultural competence is a personal characteristic, particularly if it is a characteristic that is normally (and thus naturally) distributed, then the characteristic is by definition relatively immutable (like intelligence). So what is our rationale for claiming that our training or coaching can help people to get more of it? Doesn’t this throw us in with discredited education programs that claim to raise your kid’s IQ? It may well be that the only supportable claim we can make for these kinds of measurements is that they can guide the selection of people with more IC for tasks that demand it.
Of course, the same criticisms that have plagued IQ as a selection criterion would also be justified for IC. What if there are various kinds of intercultural competence, as we now more or less accept about the existence of different kinds of intelligence? What if our measurements are biased in content and/or process (as they certainly must be), meaning some people are systematically privileged or disadvantaged by the measurement? Or, what if intercultural competence is not normally distributed? In that case, it wouldn’t be possible to rank order people in terms of IC at all.
The road less traveled here is to try to make our observations of intercultural competence more sophisticated. Unlike the positivist cast of assuming the existence of IC characteristics, we could bring a more constructivist view to the behavior of intercultural competence.
We would need to begin by assuming that “intercultural competence” is not a thing, but a product of our observation. The purpose of organizing our observation in this way is to understand how people adapt to cultural difference and to help them do it better. To continue the constructivist approach, we could assume that people get better at adapting to cultures as a function of how they organize their perception and thus their experience of cultural difference, and that we can help people do that better. (I’m sure you all recognize the DMIS.) However, if we reify peoples’ behavior into a measurable thing, particularly a normally-distributed thing that acts like IQ, we lose our rationale for guiding development. In other words, this is another case where we are shooting ourselves in the foot.