Integration, statistics and mistakes of logical typing

Are Muslims integrating or not? Are they loyal to their non-Muslim nations or not? Do we have an enemy within? Many questions for many answers. Normally mass media and in particular newspapers are the main sources of these questions and surveys and polls are the answers. Many questions and many surveys, more or less official, methodologically sound or unsound, private and public, ideological or apologetic have followed 9/11 all around the ‘Western world’. Many numbers and few words are used to convince the public that Muslims are either dangerous aliens or better citizens than the non-Muslims. A battle of opposite perspectives with only one thing in common: numbers.

The main discussion tends to be integration. Muslims are tested and re-tested about the state of their integration, even when they have been an integral part of a country for three or more generations. A recent example of this kind of exercise has been conducted by Dr Nandi from ISER and Lucinda Platt, Professor of Sociology and Director of the Millennium Cohort Study at the Institute of Education, University of London, whom have used new information collected as part of a major household survey called Understanding Society. The research, which examines how British people feel about their nationality, has revealed that people from ethnic minority backgrounds identify more closely with Britishness than their white counterparts. Among these, high is the sense of Britishness among Muslims in the UK as other surveys and statistical data show.

However, other surveys have suggested a radically different picture: a 2009 poll  showed that 32 percent of Muslim students in 30 universities across the UK believe killing in the name of religion is justified; 40 percent want Muslims in the country to be under the Sharia law; 40 percent feel it is unacceptable for Muslim men and women to mix freely; and 24 percent do not think men and women are equal in the eyes of Allah, the conclusion being that Muslims are not integrated. Another survey suggests that Muslims in the UK are less tolerant than others in Europe, at least as far as homosexuality is concerned.

Now, I do not dispute the methodology of these studies. I am sure that in the majority of the cases they tend to be statistically correct. My concern is deeper: I wonder what these tools are measuring and whether academics should engage in the ‘are Muslims one of us?’ diatribe. Before I introduce some of my theoretical observations, let me share with you what I have learnt about ‘integration’ as an anthropologist.

We need to reanalyze the word ‘integration’. The word derives from the Latin integrate, “to make whole”. If so, today the UK, as many other western countries, are certainly not ‘integrated’ countries. Yet the reality is rather different and we can suggest a better definition: integration is a political ideology aimed at the absorption, through non-violent means and through a slow process, of the minorities within the majority’s hegemonic model of lifestyle and worldview.

Nonetheless even this definition is unsatisfactory for one reason: integration is a process and not a thing, and as such it cannot be measured with one single variable. Furthermore, ‘integration’, far from being static, is affected by temporal dynamic aspects and, even more so, by space.

Hence ‘integration’ has three dimensions: social connectivity, temporal dynamics and space. In other words, integration, whatever it may mean in its multi-dimensional variables, has a minimum denominator defined by how people interact, where and when. People integrate in the place where they live and not within an idea; people integrate as part of being there in the community, people are affected by the local reality. In other words, you never integrate into a nation (which is a label and idea) but rather into communities.

This bring us to a conclusion: all people are integrated as a process and a lack of integration can only be achieved though a voluntary act of resistance. It is not a surprise that the majority of Muslims are integrated, and sometimes so well that they are affected by the same health, mental health and social issues as the majority of the British, American and Australian (just to mention countries I know) majority population.

This brings me to the first important observation about using surveys and polls to ‘measure’ the Muslim population and their ‘integration’ or lack of it, within a nation. The majority of these surveys do not measure integration as defined above, but rather opinions, often related to personal beliefs, creeds and cultural traditions. We can take for example the question about Shari’a. The question is often presented in these terms: “Do you support the introduction of the Shari’a in ….” . It is not difficult to understand that the answer will derive from the interviewees’ personal belief and from what they have to answer as “good Muslims”.

A good Muslim (or one whom wants to be perceived as such by the majority of his peers)  should believe that the Shari’a is the best of laws and should express confidence in that belief. Yet this does not mean that the same person rejects–in his or her everyday life–the British law. Questions like the one above about Shari’a  are no different to asking attendees at a Conservative Party convention if the death penalty should be reintroduced . You would not be surprised if more than 50% answer in the affirmative. Death penalty is against British (or Australian) law and values and is rejected as inhumane by the majority of the population. Should we challenge the ‘integration’ of such conservatives?

My point is that these surveys do not determine integration but rather opinions, emotions, feelings and perceptions of what people think is the correct answer to avoid cognitive dissonance. The measure of integration, if really we plan to do it, can only be understood through observation (observing people’s interactions with others, in their working places, and so on). It is actions and not words that provide an answer to the question.

Therefore, going back to Dr Nandi and Prof Platt’s study, we have to acknowledge that it only suggests  that Muslims (in reality Indians, Black Africans, Pakistanis, Bangladeshis, Middle Easterners)  have answered on a national survey, often administered by a white person, that they identify with Britishness. This shows, according to the study as well as the newspaper reporting it, that Muslims are integrated. Let me clarify some points:

  1. The survey was based on ethnicity and religion
  2. It says that Muslims (from Indians, Black Africans, Pakistanis, Bangladeshis, Middle Easterners) identify more closely with Britishness than their white non-Muslim counterparts
  3. It concludes that the expression of patriotism means that these Muslims are integrated

Again, to an attentive eye, problems appear:

  1. What kind of Muslims?
  2. Should we assume from the study that white Muslim converts identify less closely with Britishness and if so, should we, as it seems, adduce this to the fact that they are white?
  3. What is the relationship between patriotism and integration? For instance, Nazi supporters in the UK are extremely patriotic yet they are surely not integrated citizens.

At the end of the day, the survey and polls approach to the discussion of Muslim integration suffers from what Gregory Bateson has defined as a mistake of logical types:

A) No class can be a member of itself. In our case the class that is comprised of all the real Muslims in the world is, itself, an immaterial logical concept that must not be confused with the Muslims themselves that live in our streets.

B) Similarly, a class of classes cannot itself be one of the classes which are its members.That is, the class of, say, Bangadeshi, Pakistanis, Middle Easterners, or even white British (which will be comprised of the classes of Muslims, Christians, Atheists and so on) must not be confused with the class of Muslims—even though the class of Muslims is one of the classes that it contains.

C) Possibly the most relevant and important of all in our context: a name is not the thing named. Bateson says that this is the mistake of eating the menu card instead of the meal. He means that we must not confuse the name we give to an entity (“Muslims,” “Arab,” “Pakistani,” “citizen,” “integration and so on) with the real existent thing or the concept itself

C) The term “logical type” represents orders of abstraction: the class of Muslims is of the same logical type (order of abstraction or generality) as the class of non-Muslims. The class of all citizens would be of a higher (more general) type than the class of Muslims or the class of non-Muslims.

Any contradiction of these logical categories is a mistake of logical types. Unfortunately such mistakes are not mere mistakes in logic, but rather they have an impact upon on our life and policies. Next time try to read these kinds of surveys purporting to test the integration, or lack of it, of Muslims in terms of logical types and more often than not you will see they are affected by quite serious errors.

2 thoughts on “Integration, statistics and mistakes of logical typing

Add yours

  1. I really loved this article! So many answers on a lower level have been addressed and answered by your wonderful analysis. Thanks for your efforts and putting this out in a public domain.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create a website or blog at

Up ↑

%d bloggers like this: