The Ravages of Reification:

Considering the Iceberg and Cultural Intelligence, Towards De-reifying Intercultural Competence

Keynote Presentation for FILE IV, Colle Val d’Elsa, Sept. 28, 2013.


Milton J. Bennett, Ph.D.
Intercultural Development Research Institute

What I’d like to do with you today is to extend a conversation begun with a couple of blogs I have published recently, adding a specific application to the topic of assessing intercultural competence. One of blogs that I wrote in June held that was time to retire the iceberg as a metaphor for culture. There were dozens of responses to that blog, from far more people than I thought would be reading the blog. This led to a very interesting conversation that I think is related to intercultural competence. The other blog is one I wrote comparing intercultural competence to the idea of intelligence, particularly its measurement as IQ. The issue around both blogs is that of “reification” (I’ll give you the definition in a moment) and it is also the issue I believe we are dealing with in trying to talk about the assessment of intercultural competence. Basically, we need to “de- reify” the ideas of culture and intercultural competence back to some original root definition that would allow us to reestablish a more coherent approach to assessment.


Basically, we need to “de-reify” the ideas of culture and intercultural competence back to some original root definition that would allow us to reestablish a more coherent approach to assessment.


To put this talk in cultural terms, I’ll be taking a more European than American approach. Americans tend to start out very optimistic about everything and in the end they do a little criticism. Europeans tend to do the reverse; they start with being very critical and sometimes end up with constructive suggestions. Although I’m probably basically more American than European in my thinking, I’ll try here to practice something approaching the European approach.

The idea of reification in this context is attributing objective reality to a process, frequently through measurement. So for instance all of us human beings are participating in a process of defining ourselves vis-à- vis other people around us. This is the underlying idea of “identity.” But the moment we say “what’s your identity?” or “do you have an identity?” the process of generating our relationship with others becomes a thing that you either have or don’t have. That in a nutshell is reification: we objectify an ongoing process and thus turn it into a static thing.


But the moment we say “what’s your identity?” or “do you have an identity?” the process of generating our relationship with others becomes a thing that you either have or don’t have. That in a nutshell is reification: we objectify an ongoing process and thus turn it into a static thing.


Another example of reification in intercultural work is the concept of “culture shock.” Cross-cultural situations certainly generate some kind of disorientation. If they’re paying attention, people who to some extent are experiencing the world in a way that is unfamiliar are also experiencing some disorientation. However, to ask if you have culture shock is a reification of that experience, like asking if you have a certain kind of identity. Further, to assume that disorientation occurs on a U curve or a W curve in which something happens about this far into the process and then something else happens here and so on is an additional reification of the process associated with measurement. As Kay Barado and Bruce La Brack have pointed out, there is no systematic measurement support at all for these curves and we shouldn’t use them as generalizations about people’s culture shock experience.

When measurements of groups are applied to individuals, they always generate reifications. Sometimes, of course, such reifications serve us as useful diagnostic categories – a way to classify individual experiences. Such classifications need to be 1) supported by research and 2) useful for the purpose of the observation. Measurements are driven by the questions we ask. So, for instance, the original concept of culture shock generates attempts to measure a discrete kind of experience distributed among people over time; in other words, the U or W curves. Since these measurements appear to be unsuccessful, we may have asked the wrong question. Rather than asking “did you have culture shock, and when?,” we might ask “how are you dealing with the disorientation that is associated with being in another culture?” This is a different kind of question and it leads to a different kind of measurement. If we think that culture shock is a thing, then we figure out how to define that thing in such a way as to measure whether it is there or not. However if we think that people are engaged in some sort of process that involves being disoriented, then we need to inquire into the nature that disorientation and to see how it is educational or not in terms of the outcomes of the program. In other words, by staying closer to the process (i.e. reducing reification), we may enable more useful observations. I would say that this is certainly the case for “culture shock.”


Diversity is one pole of a dialectic, the other pole being “unity.” Diversity and unity need to be defined in terms of one another, like “left” and “right.” If we pull on one side or the other of a dialectic, it generates a reification.


Another example of reification in intercultural work is the way we talk about “diversity.” Diversity is one pole of a dialectic, the other pole being “unity.” Diversity and unity need to be defined in terms of one another, like “left” and “right.” If we pull on one side or the other of a dialectic, it generates a reification. So if we talk about either “left” or “right” without reference to the other, it implies that there is some kind of independent thing that has the quality of “leftness” (sinister) or “rightness” (dexterous). Similarly, talking about diversity without reference to unity leads us to posit all kinds of qualities and implications of “diversity,” as if it were a thing. Actually, left and right refer to the process of directionality, and diversity and unity refer to the process of differentiation. For directions, the more left you go, the less right you go, and vice versa. For differentiation, the more diversity you perceive, the more distinctions you make and the unity you perceive, the fewer distinctions you make.

In general, when we reify we are neglecting human authorship. We assume that a thing is “out there” in the sense that our perception or definition of it either reflective of the thing’s reality or not. This is, essentially the positivist paradigm that I have described in my article on that subject in the book Student Learning Abroad. In practical terms, even if we philosophically realize that humans have constructed a concept, we easily forget that we made it up and treat the construct as if it were a thing.

In intercultural work, the construct of “culture” is the most significant example of reification. Whenever we talk about culture as if it is a thing, we are neglecting our authorship of the construct. We forget that seeing cultural differences is a function of employing observational categories that have been constructed by us human beings. In forgetting our authorship, we may treat culture as if it is an element of reality that predates human existence, odd though that is when stated (how could there be human culture without humans?). A somewhat more constructivist assumption is that we in fact are creating the observational categories but their purpose is to reflect an underlying reality. Korzibski commented on this in General Semantics when he said that “the map is not the territory,” meaning that we should be careful not to mistake the indication of something for the thing that it’s indicating. In the case of culture, the implication of General Semantics is “when we talk about culture we’re not talking about the thing of culture, we’re talking about a map of culture. Let’s not mistake the map with the territory itself.”

Thinking of culture as maps is an interim way of thinking, however, since it still assumes that there is an underlying stable territory of culture. A more radical position is closer to Gregory Bateson’s idea, which I’m calling these days “the map is the meme.” Bateson thought that the idea of culture was constructed by human beings in order to reflect on their group experience. This is similar to how Maturana and Varela describe the “cultural level of perception” – experiencing our descriptions of reality as part of the reality that we experience. In this view, a group’s particular ideas about itself become memes that perpetuate their shared experience; in a kind of self-fulfilling prophecy, the way we describe ourselves culturally creates the “culture” that we are describing.

This idea self-constructing culture is also a kind of update of Marshall McLuhan’s idea that “the medium is the message.” Media (maps) are not neutral. Television and social media are not simply neutral channels through which things happen. They themselves are generating a construction of reality, which is particularly obvious in social media. But like all reifications, we forget our authorship of the media and assume that the reality that they create was there all along. A good way to see how maps become memes is to look at some actual maps. Here is the famous map that includes the phrase “there be dragons.” (slide) This map basically defines the world in terms of the described territory and says “as soon as you get out of this defined area, you’re going to run into dragons”. This is not just a map—this is a definition of reality. Those of us who live in northern European and American cultures are used to seeing the world oriented with north up (slide) and we are surpised to see a map with south up (slide). Through maps we define a reality that places ourselves in the center.


Those of us who live in northern European and American cultures are used to seeing the world oriented with north up and we are surprised to see a map with south up. Through maps we define a reality that places ourselves in the center.


Here is the next step: we can use metaphoric maps to generate a similar reification. An example of a metaphoric map is the iceberg simile for culture that I mentioned at the outset of this talk. This metaphoric map is not a neutral thing. When we say culture is “like an iceberg” we are defining a reality – creating a kind of emotional experience. What is that experience? (slide) The sinking of the Titanic! The idea embedded in metaphor is culture is danger, it is mystery, it is a thing hidden below the surface. We can describe the top 10%, frequently in the familiar “big C” cultural terms of cultural institutions. But below the surface, in the 90%, there be dragons. And here’s the commercial: we interculturalists can help you navigate around these dragons. We start by classifying some of them – e,g. nonverbal behavior, communication styles, cultural values – implying that we have the keys for you been able to avoid these dangers. We continue by providing particular contrasts between your own internal cultural dragons (or cultural baggage, as it is more commonly called) and the heretofore unknown other dragons. In the end, the dragons are trained to engage each other synergistically.

Joking aside, the iceberg metaphor does not create an emotional reality of cross-cultural discovery and contextual agility. On the contrary, it generates a dangerous and static territory where people must tread with caution. Insofar as this map has been reified, it implies that culture really is that territory. When the map becomes a meme (and I think it has, judging from many of the responses to my blog) then the metaphor is guiding our collective experience. As professionals, we naturally gravitate toward other metaphors and approaches that “help navigate” cultural differences, and our clients naturally demand more specific directions (tips and recipes) on how to avoid the dangers.

Since we are constrained by language to use some kind of metaphor in describing phenomena, the question should be, “does this metaphor set up a reality that makes it likely that students (or other clients) will want to develop this interesting competence we call intercultural communication? I don’t think so, even though (and perhaps because) the metaphor generates an immediate sense of understanding. It is likely that the response to the iceberg metaphor is not just the Titanic sinking. It elicits a deeper sense of hidden unconsciousness a la Freud: the id is dangerous territory that needs to be carefully navigated by the ego. The Freudian concept of unconsciousness is certainly memetic in European and North American cultures, and so its elicitation has a powerful sense of reality. But the reality is one of conscious (cognitive) control of dangerous (emotional) forces. This may be what leads our clients to demand cognitive recipes and lists of do’s and don’t as they encounter unfamiliar cultural territory, and in may be what leads us professionals to collude with them in taking an overly cognitive approach to training. (By the way, “cognitive” does not mean “didactic.” We can be use very experiential techniques to provide people with cognitive tools.)

By the way, I prefer the river metaphor for culture, since we have to compare culture to something. In a river, the water is always flowing. It’s never the individual water molecules but it’s the same configuration of water, constrained by its banks but at the same time slowly carving those banks in different ways. The river changes over time but at any one moment it’s fairly stable. Lots of things you can say about rivers are similar to what you could say about culture, and in any case it’s a more inviting idea to go take a trip on this river than it is to crush into that iceberg. I am criticizing the metaphor and exploring alternatives. And so, what we probably should be doing is thinking about how to use metaphors in a way that is consistent with the purpose that we have for intercultural learning.

Whatever the metaphor we use, the idea is to not unnecessarily reify the thing that were talking about, particularly in a way that generates a meme or a feeling that is not what consistent with the outcomes we are seeking. So, if we want people to approach culture as more of an experience than as a thing, we need to de-reify the idea of culture as a thing or a collection of things. To this end, we ought not to use static metaphors, and we shouldn’t use definitions such as “culture is the collection of values, beliefs, and behaviors…). Further, we should be extremely clear about measuring aspects of culture. As I’ve said, one common way that reification is generated is through measurement. There is this tendency from a positivist paradigm to assume that if we measure something, than what we measure is a real thing, as if we were not making up the observational category of the measurement. Consequently, if we measure a “cultural value,” we tend to assume that there really are “cultural value” that generate the variation in measurement that we observe. We thereby forget that “culture” is an observation that we make about human behavior, specifically about the coordination of meaning and action among people in groups. We are observing what people are doing, not the classification of what they have done.

The reason why I’m dwelling on this issue is that when we talk about intercultural competence we have to go even deeper than just what we mean by “competence.” We need to go into what we mean by “culture,” because when we talk about being competent in dealing with culture in some way. Are we talking about being competent in acquiring knowledge about the set of things that are associated with a particular group? No, not according to a de-reified notion of culture I just described. According to such a definition, “intercultural competence” is becoming competent in participating in another context, in this case in being able to participate in the coordination of meaning and action that works better in that other context better than it does in your own.

The constructivist idea of cultural context was stated by Berger and Luckmann in The Social Construction of Reality. Like Bateson and Maturana, they said that we are both products and producers of culture. We are born into a culture and as such culture is causative of who we are at the level of identity, but at the same time it is we who have produced that culture. If we lose sight of the dialectic, if we lose sight of the idea that we have produced the very thing that is producing us, we engage in this ultimate reification, here stated by Berger & Luckmann: “Despite the objectivity that marks the social world and human experience, it does not thereby acquire an ontological status apart from human activity that produced it.” They are saying that things don’t exist outside of your production of them, just because you forgot that you did it. “Reification implies that people are capable of forgetting their own authorship of the human world and further, that the dialectic between people the producers and their products are lost to consciousness. In this way we ironically and paradoxically can create a world that denies our existence.” This has profound implications for what kind of a world we are living in. In one world, we are victims of our own assumption of absolute reality. But alternatively, we can be on- going co-constructors of our own assumptions of how to live together.

(Audience interaction: discussion regarding use of the iceberg metaphor for culture, exploring how it might be used in a more dynamic way. Further discussion of the Freudian implications of the iceberg, stressing its focus on understanding the hidden causes for observed behavior; and exploration of a less-cognitive alternative view that lack of understanding is not lack of knowledge, but limited ability to experience the other context. Improving the ability to experience differently is a different path than improving the ability to understand hidden causes – in metaphoric terms, it is learning to sail on the river rather than mapping and avoiding the hidden iceberg. This is assuming that AFS is in the business of enabling alternative experience, not in the traditional education business of expanding cognitive categories. Of course, individuals construe the world partially through cognitive categories, so they are not irrelevant to experience. However, the “level of analysis” of intercultural work is about group experience and how to access it, not predominantly about the individual sets of categories. It is a matter of definition and focus, not absolute difference between individuals and groups.)

To move directly to the topic of measuring intercultural competence, I was struck in reading the book by Stephen Jay Gould, The Mismeasure of Man” with a parallel between the measurement of intelligence and the measurement of competence. His description of the reification of intelligence through the measurement of IQ seemed very similar to a reification of competence through its measurement. I will say a little bit about how he described the reification of intelligence and then speed forward to say how I think this is going on around intercultural competence. Then let’s see if we can de-reify intercultural competence in such a way that allows us to deal with it in the same constructive vein as we are trying to do with “culture” itself.

As Gould tells it, some French educators first attempted to identify behavior that could be used to predict which students were ready for certain kinds of education. They called the constellation of behaviors “intelligence,” by which they meant the ability to engage certain kinds of school activities. That was the original idea: a way to assign kids to the proper grade levels. And then of course people said “wouldn’t it be cool if we could measure this intelligence,” just like we are saying “wouldn’t it be great if we could measure intercultural competence?”. Aside from the practicality of rank ordering numbers rather than people, the idea of measurability lent an air of scientific sophistication to what was otherwise a sophisticated but not very sexy observation method. So of a number of existing measures were brought together and it turned out that there were scales across a number of those measures that did in fact correlate reasonably well with the readiness or predilection to engage school in a particular way. The combination of those measures was called “g”. To repeat, the g factor was the correlation of measures that in turn correlated with the behaviors that had been observed. Here comes the reification: once there was an identification of a combination of measurement categories it was assumed (by a different person) that there was a thing that was being measured and that that thing was IQ. So intelligence went from being a description of behavior to being a thing inside of people that could be measured by the constellation of measures, which generated g, which in turn was called IQ.

The other thing that generated IQ was the assumption of normal distribution. First there was the idea that some people are more ready for school than others; then came the reification of that observation into measurement categories; then came the attribution of reality to an internal “cause” for the measurement; then came an assumption that necessarily accompanies statistical measurement, which is that the quality being measured is normally distributed. Do you see what happened? We went from a purposeful observation of behavior to the assumption of a measurable quality that exists within human beings that is normally distributed in a population. This is an astonishing example of the reification of an observation – the construction of a thing. People are now being rank ordered in terms of this assumed normal distribution of a quality that has no theoretical existence outside of its measurement! In this sense, IQ is like phrenology, which is also an attempt to explain observable behavioral differences with an internal cause (different brain-area sizes). The parallel of IQ with phrenology is why Gould includes IQ an example of “the mismeasure of man”.

None of this is to say that IQ can’t be predictive, since in fact it’s related to those original observations of behavior. The original observation was that there are indeed people who are more ready to deal with certain kinds of school stuff than other people, and that those people perform differently on the “g” constellation of psychometric measures.The measurement is predictive (in a defined context) because it successfully measures people’s ability to engage certain kinds of school activities. But whether it is predictive because some people in fact have more intelligence (a higher IQ) than other people is a severely problematic, and probably completely unjustified assumption. And Gardener’s work on multiple intelligences is not the cure. All it does is to magnify the problem by looking at a variety of other reified qualities that people might “have.”

Now we can talk about some similarities in the treatment of intelligence and intercultural competence. Initially we identified some interculturally competent behavior, which usually consisted of being able to act appropriately and effectively in a different cultural context, or more generally to generate a range of behavior that is adaptive to more than one cultural context. In education, we refined the definition of competence to include the

ability to acquire transferable skills from experience abroad. Sometimes we included the ability to take a different cultural perspective, but this already is heading into dangerous reification territory. Like the educational readiness of French students, we could easily identify these behaviors in situ, or through interviewing. But, like the French educators, we were seduced by the idea that intercultural competence could be measured. And sure enough, among all the descriptive measures of intercultural behavior there appeared to be some that correlated with some aspects of interculturally competent behavior. Taken together, these measures generated a “c-factor” of intercultural competence that paralleled the “g-factor” of intelligence. The next step down the reification road was to assume that people had some set of qualities that generated the g-factor: cultural intelligence. (I use the term here in its generic sense, not necessarily as it is used commercially by Livermore and others.) Then, by assuming normal distribution of the c-factor, we generate the idea of CQ (again, generic use) and the ability to rank order people in terms of intercultural competence.

In the process of measuring competence, as was the case in measuring intelligence, we assume that the product of our measurement is the producer of the measurement. In other words, we forget our authorship of the observational category and assume that some “thing” is generating the observation. The observation that some people are better than others at adapting effectively to different cultures is reified into the assumption that people have different amounts of a quality (intercultural competence) that causes that behavior. When we add the assumption of normal distribution, the stage is set for a full-blown assumption of CQ that parallels IQ. (Although the theoretical assumptions underlying the Intercultural Development Inventory are different than those of most other psychometric tests used to measure intercultural competence, it also makes an assumption of normal distribution of “intercultural sensitivity,” and as such it equally reifies that concept compared to the original grounded theory observations.)

Here is a way that we might approach intercultural competence in a less reified way. First we must ask, “is it useful for us to professionally identify people who act more or less competently in a cross-cultural situation?“ Assuming the answer is “yes,” then we need to establish carefully what we mean by “culture” and by “competence.” We’ve already talked about a less-reified notion of “culture.” How might we de-reify “intercultural competence”? To elicit another metaphor, maybe intercultural competence is like the competence of bicycle- riding. Can you tell when someone is competent riding a bicycle? Of course you can! Moreover, once people acquire that competence, they keep it for the rest of their lives. How is intercultural competence much different than that? Some people can go into different cultural contexts and accomplish their purpose in a way that demands different kinds of behavior that they generate with an ease that seems natural, like riding a bicycle down a different road. Are these people “natural” intercultural communicators? Well, not any more than people are natural bicycle-riders. In fact, bicycle-riding demands specific training, as all of you who can ride a bike know. There is nothing natural about it; but once the skill is mastered, it feels and appears to be natural! Also, we might note, bicycle-riding is not normally distributed in the population.

But our identification of people who act competently may quickly become an exercise in finding the cause of that competence. In the case of bicycle-riding competence, we might start searching for measures that correlate with competence, thus creating a “b-factor” that correlates with the competence, thence a measure of BQ that rank orders people in terms of their capability. But that’s kind of silly, right? Most people can learn to ride a bicycle competently if they want to, and it is unclear that really good riders have anything other than good training, motivation and practice.

This is how we de-reify our concepts. We try to remember that it all starts with a choice to observe something. In the case of intercultural competence, what we are observing is a kind of interaction, not a thing or a quality. We are observing people acting in context in a way that we define as competent; we are observing what set of things they have. In this sense ware are using pretty much the same idea of competence that we applied earlier to culture: culture is not a set of things and competence is not a set of things. To say either one is to reify the concept. Then we need to avoid the idea that what we observe is necessarily caused by some internal condition such as a quality, characteristic, or trait. If we do measure some of those things, we should remember that 1) we have chosen the measurement category to reflect the observational category; and 2) if we have chosen well, what we measure will almost certainly be correlated to some significant degree with the observational category that we started with.

In summary, “culture” is an observation about the human behavior of coordinating meaning and action in a group. Everybody who is successfully socialized in a culture is “culturally competent” in that culture. “Intercultural competence” is also an observation about human behavior, in this case a kind of meta-behavior that involves coordinating meaning and action across cultures, each of which has its own coordinating system. Any assumption of causality in either case – culture or intercultural competence – tends to reify the observation by creating a “thing” that underlies the behavior. Most measurements of culture and intercultural competence do in fact assume underlying conditions, and in so doing they reify the construct in the same way that measuring IQ reifies the original concept of intelligence.

So how might we approach the assessment of intercultural competence in a non-reified way? I have already made one suggestion, which is to use non-reified definitions of the concepts in the first place. This is the idea of intercultural competence as like bicycle-riding, or more technically, as a meta-coordination of coordinating systems. Definition is the key. Once having established a non-reified definition of the phenomenon, we can look at what might be good indicators of it, and finally at how those indicators might be measured. In looking for indicators, we need to avoid the “lamppost effect.” That is the tendency to look for something where the light is best. In the case of behavior, the light tends to be best around psychological traits and characteristics. Personality theory and trait measurement are both relatively well developed (compared to communication theory and communicative competence), so there is temptation to use a constellation of such measures for assessment of intercultural competence. A related well-lighted area is that of attitude measurement. The problem is that intercultural competence is not something that we have, it is something that we do. Existing psychological measure are about how much of something (the trait or characteristic or attitude) one has, not what one is doing.


The problem is that intercultural competence is not something that we have, it is something that we do. Existing psychological measure are about how much of something (the trait or characteristic or attitude) one has, not what one is doing.


One way to compensate for the lamppost effect is to define carefully the level of analysis needed to make the assessment. As I have presented before at FILE, it is useful to differentiate individual, group, and institutional levels of analysis, where the concept of “culture” exists at group level of analysis. If we want to assess cultural or intercultural competence, it then needs to be at that same level of analysis. In this view, it is a mistake to assume that measurements at the individual level of analysis (traits, characteristics, attitudes) can be indicative of phenomena at the group level. Similarly, measurements of “leadership” or other organizational abilities at the institutional level do not work as indicators of group-level intercultural competence.

So what might be a good group-level indicator of intercultural competence? Most of my work for the last thirty years has more or less been around this question, so I won’t pretend there is an easy answer. As most of you know, I have been pursuing the idea that how we organize our perception of self/other is part of the answer, which moves into the area of “perceptual competence.” In this sense, DMIS is a model of how people organize their perception of self/other in increasingly competent ways – what I call “intercultural sensitivity” – and that increased sensitivity is a good indicator of greater intercultural competence (or at least the potential for it). I won’t say more about DMIS, since it is well-known to most of you.


In this sense, DMIS is a model of how people organize their perception of self/other in increasingly competent ways – what I call “intercultural sensitivity” – and that increased sensitivity is a good indicator of greater intercultural competence (or at least the potential for it).


Another way to approach the non-reified assessment of intercultural competence is to emphasize outcome rather than methodology. Too often we approach assessment in terms of what kind of measurement we will use, rather than what a competent person can do. For instance, at AFS you first need to establish what you want people to look like at the end of the learning process, and then you can define some ways to assess how close participants have come to that ideal. Many such assessments might be rather simple forms of systematic observation. I know that we agree on this point, but we tend to forget this simple guideline in the flurry of commercial competition. In the marketplace, clients often don’t want a discussion about outcomes, they want an easily-described and uniform assessment product. If you don’t supply it, one of your competitors will. So we end up selling a product rather than the outcome, a measurement rather than an observational definition. When this happens, we are both engaging in reification and contributing to its memetic status. I commend AFS for having resisted this trap in the past, and recommend that it continue to do so for the greater good of both the organization and the field of intercultural learning.

The Ravages of Reification:

Considering the Iceberg and Cultural Intelligence, Towards De-reifying Intercultural Competence


Milton J. Bennett, Ph.D.