Event Transcript: Global Restrictions on Religion

Today we are releasing the findings from our newly released report, Global Restrictions on Religion. The study is part of a larger effort that we call the Global Religious Futures Project. It is jointly funded by The Pew Charitable Trusts and the John Templeton Foundation. The project aims to provide solid information about religion and its role in society around the world, and to that end we use a variety of methods, from survey research to demographic analysis to the coding project that we’ll be discussing today.

Our presenters are the two people most directly involved in producing this report – Brian Grim, who many of you know is a senior researcher here at the Pew Forum and the principal investigator for this research. Alan Cooperman, our associate director for research, served as the lead editor for this report. So he and Brian are the two people who are most conversant with what’s in there.

Our format here is going to be very, very simple. After Alan introduces the report, Brian will discuss some of the main findings. I’ve given them a total of about 20 minutes, and I will enforce that, Brian, be aware, because we want to get to your comments and questions.

BRIAN GRIM: That’s 20 minutes apiece? (Laughter.)

LUGO: No, it’s 20 minutes total, as I said, so you two divide it however you want but it’s 20 minutes. Now, before I turn things over to them, I’d like to mention that this meeting is on the record and it’s being recorded so that we can post a summary of the conversation on our website. If you wish to go off the record for whatever reason, please indicate that, and we will be sure to honor your request. Alan, if you could start things off for us.

We began with the recognition that religious beliefs and practices may be infringed on not only by government actions, policies and laws, but also by social groups, organizations and individuals. And, indeed, in some places in the world, social hostilities may have much greater impact than government actions do.

We developed a series of measures to try to gauge the levels of both kinds of restrictions – government and social. We phrased these measures as questions, such as, “Is public preaching limited by any level of government?” “Is there harassment or intimidation of religious groups by any level of government?” And on the social side, “Is there mob violence related to religion?” Or, “Did individuals use violence or the threat of violence to enforce religious norms?”

In all, we have 33 questions. Luis hates the number 33, by the way. He kept saying, couldn’t you make it 30 or 32? So I have to point out now to Luis that 33 is a totemic number that even appears on Rolling Rock beer bottles. If things go well today, Brian and I will do some further investigating on that.

Seriously, there is no magic in the number of questions. That’s just the way it turned out – 20 solid measures of government activity, 13 of social activity. And to answer these questions – and this is one of the really innovative aspects of this study – to answer these questions, we turned not to the general public or even to a panel of experts, but to a set of primary sources.

We assembled a team of researchers under Brian’s supervision, who combed through 16 widely cited, publicly available documents, including reports on international religious freedom by the U.S. State Department, the U.S. Commission on International Religious Freedom, the United Nations Special Rapporteur, the Council of the European Union, and numerous other governmental and nongovernmental organizations, including Human Rights First, Human Rights Watch, the International Crisis Group and the Hudson Institute.

Brian’s team coded – that is, categorized and put into a database – all of the specific factual information that they could find in every one of these sources that would help to answer each of the 33 questions. Now, you’ll find lots of detail on this in the extensive methodology section of the written report, which we’ll hand out at the end of the meeting, and suffice it to say right now that we spared no time or expense in applying the best practices in social science research, including double-blind coding and inter-coder reliability ratings.

And Brian and his team independently coded two separate years of data from mid-2006 to mid-2008 in order to establish a firm baseline for what we hope will now be annual reports. Since many of our primary sources are reports on religious freedom, some of you may be wondering, why is the Pew Forum’s report titled Global Restrictions on Religion? For a place that generally does a good job of presenting its research to the public, isn’t that about the dullest title imaginable? Why not “Religious Freedom Under Siege,” or something like that?

Well, the short answer is that we set out to do a quantitative study, and freedom, which is commonly defined as the absence of restraints, is difficult if not impossible to measure. So we inverted the concept, and instead of trying to measure the absence of something, we’re measuring the presence of restrictions. This is not just a semantic matter; it has significant implications, and I want to spell a couple of them out, particularly for this audience today.

One is that when people think of religious freedom, they may have in mind the degree of religious dynamism, diversity and expression in a country, and the Pew Forum has tried to measure those in previous studies, particularly in our benchmark U.S. Religious Landscape Survey. This report focuses on the other side of the coin, on the impediments to religious beliefs and practices.

In addition, when people think of a, quote, “report on religious freedom,” unquote, the first thing that may come to mind is a report aimed at naming and shaming countries of, to coin a phrase, “particular concern.” That’s a perfectly legitimate thing to do, but it isn’t the goal of the Pew Forum study. We are not attempting here to determine whether particular restrictions are morally justified or unjustified, whether they are legal or illegal under international law, why they have arisen, or who is to blame for them.

Of course, we also don’t intend to sanitize restrictions on religion either. The report, as you will see, talks about violence, deaths, torture, detentions, displacement from homes, and I don’t mean to suggest that we view such actions with equanimity. But I want to stress that in keeping with the Pew Research Center’s fundamental ethos as a nonpartisan, nonadvocacy fact tank, our goal here is simply – or maybe not so simply – to measure restrictions on religion in an objective, transparent and reproducible manner, and we’ll leave it to others, including many of you in this room, to do what you do so well, which is political and social advocacy and policy formation.

One final point: When Brian and his team combed through the primary sources of data for this report, they looked for specific, factual information about government laws and policies and for detailed, verifiable accounts of social incidents, complete with names, dates and places. They did not code or record the opinions or overarching judgments of the sources.

We think this is one of the strengths of the study, but it does have an unfortunate side effect. We had to exclude North Korea from our indexes. As we note in the report, the underlying sources are in virtual unanimity that North Korea is among the most restrictive countries in the world. And as many of you know, the flow of defectors from North Korea in recent years has produced a lot of good scholarship.

But still, as our sources themselves acknowledge, there is a paucity of the kind of specific, timely data that we relied on for this study, and we hope that in the future that may change. The study we’re releasing today is just a snapshot in time. We want to repeat it in future years, which will give us the ability to capture changes and trends. It may also allow us to improve our measures, and we welcome your suggestions on how to do so.

So now that I’ve cleared our institutional throat and maybe whetted your appetite and prepared to quench my own thirst later, I give you the principal researcher on this long and complex project, the indefatigable Brian Grim.

Also at the start I’d like to acknowledge the great work that the Pew Research Center and many staff here have put into this report, helping get it out to the world. Researchers like myself don’t mind just holing up and doing research, but to get it out to the world took a whole team of people – editors, graphics designers, administrators all up and down the Pew chain of command – to review this report because it is a first of a kind.

It’s a new methodology for the Pew Research Center, so I’d like to express my gratitude, and Alan came in – he came in not quite as a fourth-quarter quarterback, but he pushed this across the goal line. There has been a lot of work done through the years, and so I appreciate all the great help that’s been given. And that doesn’t count against my minutes, my acknowledgments. (Laughter.) Reset.

It is a humbling pleasure to now share some of the main findings of the study. At the start I should mention that Article 18 of the Universal Declaration of Human Rights, one of the foundational documents of the U.N., summarizes the understanding of free religious practice that we’ve used in this study. Article 18 states, “Everyone has the right to freedom of thought, conscience and religion. This right includes freedom to change his religion or belief, and freedom, either alone or in community with others and in public or private, to manifest his religion or belief in teaching, practice, worship and observance.”

Our study finds that 64 nations, or about one-third of the world today, have high restrictions on religion, but because some of the most restrictive countries are very populous, that means about 70 percent of the world’s population lives in countries with high or very high restrictions on religion.

I’d like to now unpack what goes into these numbers. As Alan mentioned, we look at two dimensions of restrictions on religion. The first is government restrictions and the second is social hostilities involving religion. I’ll use this presentation to unpack these and give some examples of the measures that have gone into the scoring.

Looking at some of the countries scoring very high on government restrictions – Saudi Arabia topping the list, Iran, Uzbekistan – many of the governments of these countries tend – if you look down the list, you can recognize they tend to support one or more religious groups or perspectives on religion above others, and they often make efforts to silence competing perspectives on religion, including through the use of force. One thing to note is that there is just one country that appears on both lists – Saudi Arabia. As some of you may know, its score is driven up largely because they have a semi-official but really a social police force that enforces a strict version of Islamic law, Shariah law – the muttawa. They stand on street corners and act both with semi-official authority and as a social group, enforcing religions norms within the country.On the right hand are the top 5 percent of countries that score very high on the Social Hostilities Index. If you look at this list of countries topped by Iraq, which is logical, followed by India, Pakistan, Afghanistan, you will see that violence results often from tensions between religious groups, not just across civilization divides, so to speak, but within religious groups. So in Iraq, it’s sectarian violence, and Pakistan, similarly, some sectarian violence. And in Afghanistan, also, you see tensions within Islam within their own country.

The highest overall restrictions are found in countries where there are both high government restrictions on religion and high social hostilities involving religion. In this chart we’ve mapped social hostilities going up the chart and going to the right are higher government restrictions.

If you look up in the top right-hand corner, these would be the countries where you will find the highest restrictions on religion because they’re coming both from government and from forces within society. Pakistan, Indonesia, Saudi Arabia – these are just the 25 most populous, but if Saudi Arabia were mapped here, it would be right above Egypt in this area, Egypt and Iran.

So among the largest, most populous countries, those would be the most restrictive, in our measure. But government restrictions and social hostilities, while they tend to move in tandem, there are some notable exceptions. You can see China and Vietnam have extremely high restrictions on religion from the government but relatively fewer restrictions coming from people in society and groups in society.

On the opposite end of the spectrum, you can see India up on the top part of the chart, where social hostilities tend to be higher than government restrictions, though both tend to be high. Nigeria is another example of a country where you have higher social hostilities than you do government restrictions on religion. So this map presents a way to understand the pressures within a country on religious groups. It’s not all the same. It’s not all restricted by governments. It’s not all restricted by social groups.

Then in the bottom left-hand corner of the chart, you’ll see a cluster of countries. There are only two that are low on both measures among the 25 most populous – Japan and Brazil. The U.S. falls into the moderately restrictive category in terms of social hostilities, and that score is driven by frequent, religiously biased hate crimes. Each year law enforcement officials report about 1,400 religiously biased hate crimes in the United States, covering nearly all 50 states. So the U.S. score is a bit higher there, and I’ll say something more about the U.S. in a bit.

Now I’d like to unpack a bit more what went into these scores. They’re not magic numbers; they didn’t come out of a black box. They came by looking at individual questions. I’ll run through several of these, and some I’ll mention more briefly, and during the question and answer time, we can talk about examples or if you have questions on them.

But a logical place to begin, and one that many people begin with when they try to understand restrictions on religion, is whether there are any constitutional protections for religion. And, surprisingly to some, in three-quarters of constitutions in the world today, there is an explicit provision for protecting religious freedom. So if constitutions were a good predictor of what happens in society, we could stop here and say, whew, we’re done and we don’t need so many hours looking at other things. So 20 percent more provide some protection for religious practices, and only 4 percent of constitutions in the world provide no protection for religious freedom, or no provision.

But again, when we look at constitutions a bit more closely – and our team went through every constitution of the world, every constitution available – we ask a second question: Does the constitution or basic law include any stipulations that appear to qualify or overtly contradict the principle or the promise of religious freedom?

When we look at it this way, only 22 percent of countries have no qualification or contradiction embedded within their constitution. These qualifications may be as simple as, you can have religious freedom as long as you don’t disturb public morality, whatever that actually means, and who is the judge of what public – how to judge public morality?

A contradiction may be that there is religious freedom in the country or some religious practices are protected as long as they don’t contradict a certain religious law or a certain principle in the country. So in Indonesia you cannot legally be an atheist. Everyone has to subscribe to a religion. In Afghanistan no law can contradict the sacred law of Islam. That means that any other religious group has to live underneath or within those set stipulations in the constitution, which then decreases their ability to have religious freedom with respect to their own religion.

Many of the 20 questions we looked at on the government side look at specific acts that governments can take. We only looked at two questions on constitution because we found that constitutions themselves were not so predictive of the overall level of religious restrictions in a country.

So one question is, “Are foreign missionaries allowed to operate?” In just over half the countries they are. That means in just under half the countries they’re restricted in their activities by governments, and in about 6 percent of countries they’re not permitted to work. These are not just Christian missionaries; these are missionaries of any sort. So in many countries – Muslim countries or Buddhist countries – they’ll restrict any kind of missionary other than the ones that they’re favoring. It may even be a rival sect within the religion.

Another question we looked at: “Was there harassment or intimidation of religious groups by any level of government?” That’s another feature of this study that’s very different from other studies where researchers will look at large-scale violence and large-scale government actions. We looked at whether there was harassment by any level, and indeed, most harassment and most intimidation of people happens in their neighborhoods by their local police, by their local schools, by local government institutions.

When we looked at this, we found that in about one-third of the countries there were no reports of government harassment, but in more than two-thirds of the countries there was either harassment or some sort of intimidation, and in more than a quarter of the countries there was widespread intimidation of one or more religious groups within the country.

And this is a seemingly neutral requirement. In fact, nine out of 10 countries do this. They require religious groups to register for one reason or another. This may be to get tax-exempt status, or to be able to construct a building you need to have a legal presence and therefore register yourself, or to import religious literature for your congregation from abroad – a host of reasons governments can ask a religious group to register, and that seems reasonable.

However, when we start to look at these registration requirements, we find that in more than half of the cases – nearly 60 percent of the cases where religious groups are required or asked to register by a government – these requirements are used to either control their activities or in some way – in 40 percent of the cases – clearly discriminate against certain groups by allowing some groups to register and other groups not to register.

Government funding, this is a kind of favoritism of religion by the government. That isn’t another type of activity that people in the world, outside of America maybe, think of as a restriction. It’s a very normal activity. More than 80 percent of the world’s governments provide some funding or resources to religious groups, including the United States, by the way. So it seems that this is not necessarily a bad thing; this is helping to promote religion. We’re not trying to restrict religion. But as the adage goes, wherever there is money there are strings attached. You know, he who pays the piper calls the tune.

What we find is that, indeed, in three-quarters of the countries of the world this funding is provided, but it is provided in some imbalanced way. It’s provided where one religious group will receive favors that another religious group doesn’t. It seems natural that this would happen, but we view it as a restriction because it makes the playing field, so to speak, un-level. Those groups that are excluded from the favors are disadvantaged in a way that they wouldn’t be otherwise.

Another measure – again, if you’re looking at the white parts of this chart, that’s the non-restrictive area. There are only about one in five countries that ban religions. That’s one way to say it, or one in five countries ban one or more religious groups. It seems like a small number, but when you think about it, that’s a fair proportion of the world’s countries.

Many times these bans relate to security reasons stated as the rationale. It might be Hizb ut-Tahrir or some other group that’s banned in the country that has a predominantly religious nature but also a political agenda. But many times governments will justify banning these religious groups on security grounds.

Now, looking at the number of countries that have high levels of government restrictions, we look at the top 20 percent as having high. We didn’t look at our data and have an arbitrary cutoff where we say, well, these are the high ones, these are the moderate ones and these are the low ones. We took it strictly as a percent.

The top 20 percent of countries scoring the highest on our measure we said are high, and the top 5 percent are very high. Then the next 20 percent we said were moderate, had moderate restrictions. And then the bottom 60 percent we said are low because most of these countries had scores around 2.3 or less out of 10, and many countries in this 60 percent had scores of one or – very low-level restrictions.

However, in that 20 percent of countries with high or very high restrictions, 57 percent of the world’s population lives in those countries. And of course that’s driven by some large population countries such as China.

Looking at how these restrictions play out across the world, the region of the world that has the highest median average of restrictions is the Middle East and North Africa – nearly five times the median level of restrictions as the Americas. You can also look at Asia-Pacific being in second place. The situation of Asia-Pacific is interesting because you have some of the lowest-scoring countries – Japan and Taiwan have very low levels of restriction – but then you have some countries with the highest restrictions, such as China and Burma, among governments.

Europe – its median is higher than sub-Saharan Africa, so there are more restrictions in Europe than there are in sub-Saharan Africa, according to our measures. This is largely driven by countries in Eastern Europe, where numerous restrictions still remain on religion, but also Western Europe’s scores contribute to this. In many countries in Western Europe, there is an attempt to protect the society from groups that they consider dangerous or cult-like. So there are numerous policies that are restrictive in Western Europe.

Now I’ll go through more quickly the social hostilities, just to give you an idea of the questions that went into this measure. Those were a sample of the questions that went into the government restrictions. Now some that we included in the Social Hostilities Index. The first was, “Were there tensions between religious groups?” We found that, yes, indeed, in the majority of countries in the world there were public tensions. We’re not talking about tensions in somebody’s family that don’t spill out into the public but reported public tensions, and in about one in five countries, these had physical violence in numerous cases.

At the upper end of this type of violence, we see sectarian violence occurring in about one in 10 countries in the world – sectarian violence meaning that there is a repeated conflict between two opposing religious groups. This could be Sunni and Shia; this could be two more-localized religions in a province of a country. It doesn’t mean widespread civil war, but it means religious groups repeatedly and violently facing off against each other.

In our measure on social hostilities, we looked at hostilities over specific events, such as whether there were hostilities over conversion. We found in about one-third of countries that there were public tensions over conversions, and in 16 percent of countries we found that there was physical violence related to someone converting from one religion to another.

This is a measure that social scientists like. It was one that those who aren’t in social science sometimes have a hard time getting their minds around, but I think – I’ll go through this very quickly. There’s a concept in sociology called social movements and that these have power within society, forces within a country that can dictate attitudes and opinions. We looked at whether or not there were organized groups that tried to create movements that use force or coercion in an attempt to dominate the public life of a country by opposing other groups.

Some simple examples might be neo-Nazi groups that are motivated by religious or even ethnic bias in Europe, or the Ku Klux Klan in the United States, which has a religious motivation that spills out into ethnic and religious violence, or extremist movements in Asia and the Middle East – organized groups to forcefully institute Shariah law, for instance. We found that in about 66 percent of countries these types of movements were active, and in about 26 percent of countries at the national level.

Another finding is that in the majority of countries, religious groups themselves tend to oppose the activities of other religious groups and try to corner the market, so to speak, on religion and keep other religious groups out. In about 53 percent of countries we found this. On a large scale you can see it in Russia with the Russian Orthodox Church advocating for limits on new religions coming into the country.

In about one in four countries, there are groups or individuals that have been reported to try to enforce religious norms. This may be honor killings – religiously motivated so-called honor killings, where they’re targeting females who may have gotten pregnant out of wedlock or married someone from another religion and they feel it’s their religious duty to kill these women or at least shun them in some serious way. We find that happening in about a quarter of countries.

And then finally, we looked at terrorism and war as part of our social hostilities measure – religion-related terrorism and religion-related war – and we found that in only about one in 10 countries – or, put the other way, in one in 10 countries, religion-related terrorism resulted in casualties during the period we studied. Additionally, we found that in 22 percent of countries there were religion-related terrorist groups active in some way, either fundraising in a country or recruiting members, and so the United States would receive a score at that level on this question.

This is similar to the government restrictions. Looking at, again, the 20 percent of countries at the high level, almost half of the world lives in those 20 percent of countries – 46 percent in countries with high social hostilities involving religion.

This looks somewhat similar to the chart that broke out government restrictions by region of the world, but there are a few differences. One is that the median score in the Middle East and North Africa on social hostilities is about seven times higher than the Americas, instead of almost five on the government restrictions. So there’s a much greater gap in terms of social hostilities between the two. And then also in Asia and Pacific, Europe, and sub-Saharan African, those three regions have somewhat similar scores, but there are some significant differences.

If you notice, Europe has a shorter bandwidth, so to speak. That means that the highest level of social hostilities reached in any country of Europe is only at the median level of the Middle East and North Africa, whereas in Asia and Pacific, though in general there are overall low levels of religious hostilities in society, there are several countries with extremely high levels of religious hostilities, Pakistan and India being some examples. And also, in sub-Saharan Africa again, though there is an overall low median score, there are some countries with very high levels of social hostilities, such as Somalia and Sudan.

Finally, I’ll end with this slide that will test your eyesight. The one thing that I’d like to just point out in this slide is that if you were to draw a regression line or a line that sees how close each country comes to the median score going up the page, these two measures – social hostilities, government restrictions – tend to move in tandem. So where you have high restrictions on one, you tend to have high restrictions on the other.

The last point that I will make is that in the low area – these are, again, the 50 most populous countries – there are less than a dozen countries that fall into the low category among the 50 most populous. That means the rest of the 50 fall at the moderate or higher levels.

A trend that you can see in the countries with the highest restrictions – these tend to be either in Asia or in the Middle East – and as I mentioned at the beginning, many of these countries are attempting to promote one religion or have one religion become the religious identity of the country to the exclusion of others. So with that, thank you for your attention, and we’ll open it up for discussion.

LUGO: Thank you, Brian. Thank you for being such a good sport on the time restrictions that I placed upon you. Let’s hand out the reports. I think it would be helpful for the discussion if these folks had them on hand as opposed to waiting until the end because Brian and Alan may like to refer to sections of the report and you folks will want to start leafing through it. All right, so let’s get started.

PAUL MARSHALL, HUDSON INSTITUTE: Alan and Brian, thank you very much indeed for the presentation report. It’s really excellent. In terms of my own judgments of seeing where countries are situated, most would fit my own judgments but some wouldn’t, and I think that may be connected to sort of a methodological question, the question of intensity. Are there restrictions or aren’t there, and then how strong they are?

One of my examples would be, perhaps, Indonesia, which in terms of its social restrictions scored fairly high, higher than I would have expected on your scales. Now it is illegal in Indonesia to be an atheist, but generally people don’t make a – it’s a don’t ask, don’t tell country. (Laughter.)

I know Indonesian atheists, and they would say, well, don’t go and stand on the street corner, bounce up and down and say, I am an atheist; religion is junk. You might have problems. But nobody is going to chase you; whereas in Saudi Arabia if you’re an atheist, you’ve got problems. So both are illegal but one is, in practice, a worse situation.

I was thinking that, too, in terms of Malaysia, of your government restrictions. Again, that seemed higher than I would have – I don’t mean I’m right; I just mean it was higher than I would have expected. And again, I wonder – in terms of Malaysia, you’ve had this sort of proliferation of anti-conversion cases so that there are strong restrictions there vis-à-vis, say, a country like Egypt, where there is no explicit law forbidding conversion but they’ll use other laws to stop that.

But I think what strikes me in Malaysia is I don’t like those laws, but they do deal with them in courts. People go to courts and they appeal and fight, and the people having the appeal go home. They don’t end up dead. In Pakistan your worry isn’t the court; you’re going to be killed before you ever get to the court.

So Malaysia has stronger formal restrictions and they’re bad ones, but there’s actually a real legal process, which is better than some other countries. So if you have any response concerning those two particular countries but just dealing with the problem – I’ll just call it intensity of restriction rather than fact of restriction.

Then we also track within Indonesia whether or not that’s proliferating. So though one province has now adopted some higher restrictions on religion than others, we’re seeing that spread to various communities. There are reports of hate preaching in some of the mosques that may not have been there five years ago or 10 years ago.

Some of these measures are sensitive to current events that are happening in Indonesia, and especially local events. The reason that we make the assumption that if there is a severe problem in one part of the country, the whole country will get this score is that at the national level, the government has the opportunity to review it. They have the opportunity to come in and correct what would appear to be higher restrictions than the rest of the country, and at this point the federal government hasn’t stepped in. Instead, it gives the green light.

That would be part of what’s driving Indonesia, and then the government itself has given semi-official or maybe official recognition to the al Ummah, the body of imams in the country. They issue things such as fatwas against Ahmadiyya being a heretical sect and calling for them not to be able to call themselves Muslims, similar to Pakistan. Once these proclamations were made, Ahmadiyya mosques were burned and they became targets for persecution within the country.

These are some of the things that go into our coding, and I think what you’re saying is that in Indonesia, maybe generally, people are sort of laid back about these matters and don’t ask, don’t tell, as you say, but we pick up these kind of incidents that are happening. It may, as you say, not reflect the general feeling in Indonesia in some places, but it does reflect actual events and facts that are happening on the ground.

Regarding Malaysia, this will pick up many of the restrictions within a religious community, so the fact that in Malaysia Muslims themselves are under much tighter restrictions religiously than, say, Christians – Christians have more freedom, unless they want to use the word “Allah,” but they are under many tighter restrictions. They have to abide by Shariah law, and there has been more emphasis on that in recent years.

So we’re picking up these restrictions, though in the country you may go around and find, well, there are churches and there are mosques and there are temples, and that seems to be functioning quite well, but it’s a very different situation than Thailand, for instance, where also you could drive around and maybe see some similar sites of religious pluralism within the country, but it’s whether that restrictive nature is being applied in part to some and not others.

COOPERMAN: I’ll just add one thing, which is that if you go to the online presentation of our report – and the URL is in the front of the written report – you’ll find a section that is very thick – it could be thicker than this entire report that you have in front of you – called “Results by Country.” And in “Results by Country” we provide the detail. Transparency is one of our hallmarks, as you know. We show how every country scored on every question. So that’s where you can find the detail on Indonesia and Malaysia if you want.

I have one methodological question and one larger question. The methodological question is, especially on the Social Hostilities Index, how did you account for availability of information? I mean, some countries get much closer coverage by the media or by outside groups than others. I noticed just going through the list, Belize, San Merino and Tonga all scored zero on social hostility.

It is not your fault that there’s probably not a whole lot of reporting on what’s happening in this area in Belize, San Merino and Tonga, so how do you take into account that fact, or how should we read this? And then the larger question is, do you, in the report, or could you now talk about the connections between restrictions on religious freedom and restrictions on other forms of freedom – press, assembly, speech, political organization? Do you care to discuss that? And thanks again for this amazing piece of work.

GRIM: On the first question, it’s an excellent question, how we address the reporting bias, especially on the social problems. And in our methodology we acknowledge that this is a problem, and we’ve taken steps to try to deal with it.

First I’d say that in terms of big problems that happen in countries, if there are big problems happening, our sources tend to pick them up. So if there is a problem in Belize, maybe it’s not so big. That would be our general assumption because these watch groups are constantly looking for problems that are happening. But at the same time, in many countries that have open access information, we’ll find more problems. So what we did methodologically to counter that problem is that we didn’t include actual counts, numeric counts, of incidents of violence that happened within countries as we did in the Government Restriction Index.

There we felt that the resources are sufficiently comprehensive that we can actually count the number of people abused by governments and it would be fairly reliable. But on the social side, we felt that it was less reliable, so instead of counting how many people – say, in China there’s less reporting than there is in France, for instance – we didn’t include those counts. Instead, we observed whether there were large-scale problems and then reported on those.

The other thing involved with the social hostilities is that we relied completely on our sources to tell us whether these events that happened were religion-related. So, for instance, this happened outside the coding period, the riots in Urumqi, in Xinjiang, in Western China among the Uighurs.

That happened outside of our coding period, but had it happened within our coding period, we would only count that as religion-related social violence if the sources made the judgment that that really involved religion as opposed to Uighurs being abused in a factory in southeast China, which triggered a riot in northwest China. They happened to be Muslims, but religion may not have been central to that riot. So everywhere that we’d make these calls on the social side, it’s not whether Muslims and Christians or Muslims and Chinese are fighting. It’s whether or not religion is some identifiable part of the hostilities.

On the second question about the correlations, the only correlation we looked at was the relationship between government restrictions and social hostilities, and we did find that there was some correlation – there were some notable outliers. We haven’t so far done a study on the correlation with others, some other academics, and if I take off my Pew hat, those are some things that I’ve studied, particularly the relationship between government restrictions, social restrictions and violence. But this study doesn’t address those. We’re providing this data and hopefully it will stimulate some more research for people to look at these things.

I have a couple of questions, also methodology related. One I think you’ve partly gotten to, which is, you mentioned the U.S. as sort of a moderate religious freedom, partly because of the hate crimes, and I know Human Rights First had done a tremendous report on hate crimes.

One of the things that they noted is that the U.S., of course, actually keeps quite good statistics on hate crimes and even breaks them down into categories, whereas countries where there is a tremendous amount of hate crimes – Russia, for instance, being one of them – the government makes no efforts, and so it’s very difficult. You have to rely on beleaguered human rights organizations to look at the press and kind of piece together a sense of how much those are taking place. I do wonder if the U.S. maybe got dinged a little bit more, partly because it really is quite good at gathering those statistics and presenting them.

The other question that I had is whether you looked at issues in your survey – and I’ll have a chance to look through it more thoroughly on your list of questions – is whether you looked at, on the government side, whether or not conversions are allowed in countries under law and whether they’re also allowed in practice, and also whether you looked at the ability of people to engage in discussion and free speech about religion, in other words, whether or not there were laws restricting the ability of people to say nasty things about other people’s religions or even question tenets within their own religions. I’m just curious whether that’s something that you looked at.

COOPERMAN: Thank you. That’s an excellent question also and I think directly related to Paul’s and E.J.’s questions. And so perhaps the best way to answer it is to refer you now to the report. Let’s turn to page 65 where we actually begin with our list of questions.

You’ll see this is now the list in the social hostilities questions, or measures. And you’ll see that we did attempt, in answer to Paul, to find ways to gauge levels or intensity, severity, on these various measures. Most are not yes or no, though some are. Let’s look at question number seven: “Did organized groups use force or coercion in an attempt to dominate public life?” You’ll see there’s a no or yes, but under yes, also at the local level, at the regional level, at the national level and so on.

And you’ll also see, in answer to Paula’s question, that in order to avoid what we would call counting error, we tried to focus on the worst kinds of consequences of restrictions, such as deaths and displacement from homes, on the assumption that those are better reported throughout the world.

We didn’t count numbers of incidents of things like vandalism or verbal harassment, only in our very first question, which is a kind of summary question, and there we just have a yes or no: “Are those things reported by our sources in these countries or not at all?” Those scores, as a portion of the total index score, would be quite small.

But if a country has a lot of deaths reported, that’s going to be picked up and may even be picked up conceivably in more than one measure. We also tried to not have too much overlap between the measures. And I’ll let Brian address –

LUGO: So the intensity question you’re saying we picked up on two counts. That is, within each question there are degrees, right? But also we count on redundancy to get at the other issue, right?

COOPERMAN: Redundancy particularly in the worst types of violence. What you’ll see is particularly redundancy in deaths and displacement from homes.

GRIM: Could you repeat the first half of your second question? You asked about whether our measures pick up any limitations on free speech about religion, and then you had a first part to that question.

SCHRIEFER: Oh, religious conversions, whether or not they’re allowed, whether people are allowed to actually change their religion.

GRIM: Oh, yeah. One of our measures in here on page – maybe Alan will find it.

COOPERMAN: It’s in the government restrictions.

GRIM: Yeah, we have it on the social but we also have a parallel question on whether governments – let’s see, on page 56, question seven. “Is converting from one religion to another limited by any level of government?” And we find that in about 18 percent of countries, or about one in five countries, it is limited in some way.

For the second, we pick up whether or not there are any laws that inhibit the free practice of religion. In one of our questions, I think question number three on page 55, we would pick up these types of restrictions in this question, whether or not – religious freedom, which we view as to be able to have free speech about religion – so if there were these reports.

But I would say that many of the reports that we coded from the 2006 to 2008 period weren’t reporting on this as a major topic. It’s become a growing topic in recent years, so it’s sort of contemporary, and the folks here from the State Department’s International Religious Freedom Office may be looking at that for their future reports as a category to report on.

COOPERMAN: Also GRI-5 and GRI-8.

GRIM: Yes, Alan points out also that GRI-5, which is page 56: “Is public preaching by religious groups limited by any level of government.” That would be the ability to preach and offer a critique of another religion or a criticism. If that appears, then we would pick that up.

And question eight is religious literature or broadcasting. That tends to be where those kind of restrictions would come into play. So we would pick them up, though we don’t have a particular question on that.

The section begins by saying, “It is important to keep a few caveats in mind when reading this report. First, because freedom – defined as ‘the absence of hindrance, restraint, confinement or repression’ – is difficult if not impossible to measure, the Pew Forum’s study instead measures the presence of restrictions of various kinds. The study tallies publicly reported incidents of religious violence, intolerance, intimidation and discrimination by governments and private actors.” So that’s the first caveat.

Then if you go down to the next paragraph, the first sentence reads, “This study does not attach normative judgments to restrictions on religion,” which is to say it does not attach normative judgments to reported incidents of religious violence, intolerance, intimidation or discrimination, just putting those two things together.

And that highlights the problem. Or to put it slightly differently, since the presumption of most people is that liberty is a good thing and more liberty is preferable to less liberty in the area of religion and other things as well, then it is very difficult to read the word restrictions as anything else than normatively loaded, in a very particular way. And just to highlight the problem, let’s go to the United States, okay, and let’s take one of your questions – GRI-18 that appears on page 59, having to do with registration for various purposes, such as taxation.

Of course, we do that in this country, and that leads to cases like the famous Bob Jones case, where, in fact, tax exemption was refused to an organization that practiced racial discrimination. It was a religiously based organization that practiced racial discrimination in its admissions policies and other policies as a matter of faith. Most people think that case was rightly decided and is normatively defensible, but it counts as a restriction.

So how do you put together the value-neutrality of the social science portion of this with the cultural environment of preference for religious liberty that will influence the way this report is interpreted? In other words, how do you block the inference that all restriction is bad?

LUGO: Excellent question.

COOPERMAN: Let me tackle it first. And first, just to say, as a factual matter, on question GRI-18 the United States is in the “yes, but in a nondiscriminatory way” category.

GALSTON: But the U.S. discriminates against discrimination.

COOPERMAN: In the question of, “Does the United States ask religious groups to register for any reason,” the answer is yes, but it’s coded in our coding as doing so in a nondiscriminatory way.

GALSTON: Okay, well then, you’ve built in a normative distinction between valid and invalid forms of unequal treatment of religious groups.

COOPERMAN: No.

None of our sources said that, and I’d say that’s really a chief limitation of our study, but it’s also a methodological strength because I think if we would go through most of these questions on the United States, we would be able to say, well, yeah, here, here, here, and we would answer yes to many of them, and the U.S. may be at the very top of the scale, partly because we care about this so much in the United States and so much attention to even small problems is given.

That’s methodologically why we confine ourselves not to doing any theorizing as we code, but merely looking at our sources and taking what they’ve culled from their data to say, here are the big problems in the country. I’m not avoiding your question, but that’s the reason the U.S. would only get a “nondiscriminatory way” response on that one question.

GALSTON: But just to sharpen the question by returning to your own text, to say in one breath that the study tallies publicly reported incidents of religious violence, intolerance, intimidation and discrimination, each one of those words normatively loaded, and then to say in the next breath that the study doesn’t attach normative judgments to restrictions on religion – it’s a little hard to put those two things together for the ordinary reader.

However, in this report we do not go through and label individual restrictions in particular countries as being justified or unjustified, nor do we attempt to say whether they are allowable under international human rights law or not. As you no doubt know, some restrictions on religion may be allowable under international human rights law. We do not attempt to say what accounts for these restrictions, that is, what historical, demographic, cultural, religious, economic and political explanations might exist.

In particular countries social hostilities may arise from the actions of forces external to that country, and we are not saying in any sense that a country having a high level, or having particular levels of restrictions or particular levels of social hostilities, that in any sense that’s a kind of a blame on that country. Let’s take a country that has been hit by terrorist attacks from outside. That might raise the level of social hostilities related to religion in that country, so we’re not saying that country is to blame for that.

Now these are all very excellent questions and should be addressed, but they’re not addressed by us in this report. And we, as a nonadvocacy organization, would have difficulty in some aspects of dealing with some of these questions, particularly questions of, are these things justified, not justified? That leads us into policy areas. So we have attempted, as best we can, to measure the levels of restrictions. By saying non-normative, I don’t mean that there aren’t some values built in, especially at the high consequences of these actions. But our goal is not a normative naming-and-shaming report, which is, indeed, a perfectly valid thing for organizations to do. It’s just not what we’ve attempted to do here.

LUGO: If you look at public opinion around the world on issues related to this, I think most Americans would be quite surprised at what they would find. In places like Russia, for instance, the recent report from our colleagues at the Pew Global Attitudes Project and their views on religious liberty – it’s very different than the way most Americans would look at it, and people place different normative value judgments on it.

As you know, Bill, with your background in political theory, increasingly for us in the United States and in the West, the notion of rights and liberties has become very attached to individual rights and liberties. And there is, as you also well know, a competing understanding of rights and liberties that are more communal in nature. So it’s not surprising to me, when I look at global public opinion, that in certain societies where the understanding of rights is more communal in nature, people have much higher support for restricting the rights of individuals, including on religious matters.

So there are larger political/philosophical issues lurking here, and partly because of that, but partly because it’s in our DNA here at the Pew Research Center, even though we’re dealing with an inherently normative topic, we try to be as non-normative and dispassionate in the way we treat it as possible. But you’re right; there are some very fine lines there. No question about it.

You pointed out at the end of your presentation, Brian, that governmental restrictions on religion tend to go in tandem with social hostility, which is really very, very interesting. And it’s interesting, too, in part because we look around the world and we see lots of evidence of a surge in religious dynamism, religion’s public influence. There’s a kind of religious resurgence globally.

So it’s interesting that even though you’ve got lots of government restrictions on religion, government restrictions on religion don’t seem to be very good at restricting religion. But they do seem to be very good at fomenting social hostility about religion. I’d like to hear you unpack the causal dynamics of that relationship. Why does there seem to be a correlation? I know that may take you a little bit away from the factual base of the report, but I’d like to hear your best analysis of why there is a relationship.

But then there’s a second part of my question, which is, there do seem to be some striking exceptions to that relationship, as you indicated. There is a regression line, but there are a striking number of cases off the line. Paul mentioned Malaysia earlier. Malaysia is among your topmost restrictive countries with respect to government regulation, but it’s one of your lowest countries with respect to social hostilities. It’s at the very bottom, I think, or close to the bottom, in Asia. So how do you explain cases that are off the line?

GRIM: Well, on the first question, I’ll have to pass a bit. But I have done some work that’s been published in The American Sociological Review that addresses the very question that you’ve asked, so if I could direct you to that – and I know you have a copy of it. If anyone else is interested, you can look at that.

LUGO: But let’s not be so coy here. I mean, if you’re looking at correlations, Tim – and this gets back to E.J.’s question – we’ve done this. You know, we’ve peeked. Are these things highly correlated, let’s say, to freedom of the press, to the level of democracy and all that? I think you were fair to say that you will find many correlations there. But the key there is, what is the causal effect, and that’s why we hesitate to go further in this report because that’s sort of the second wave of analysis, where social scientists then take a variety of factors and test it with not just correlations but regression analysis to see what may be driving things here.

And this gets to your question of exceptions. We don’t just want to see what’s the causal connection but how strongly predictive this is of that because you’re going to have outliers in any of this analysis. I would say that’s a second wave. You first have to have good, solid data on freedom of the press and levels of democracy and all that. We think this will provide a good baseline of data on the religion restriction side, and then we can begin to do that kind of analysis. But obviously that goes way beyond what we’re attempting to do here. But it’s a very important and legitimate question.

SHAH: If I may follow up, the report makes some interesting suggestions about why some cases are off the line – for example, why you have a case like China, where you have extremely high levels of government repression but very low levels of social hostility. I think the report suggests, well, it’s natural. If a country has an extremely effective, tyrannical government, it’s relatively easy to prevent the manifestations of religious hostility and tension. You can keep the lid on more effectively. Do you have other things you could say about other cases that seem to fall off the line? I’m not asking you to do high-level causal interpretation, but –

GRIM: I think the other thing we’ve alluded to or mentioned in the report is that in the other case, where there are higher social hostilities than there are government restrictions, that these tend to be driven by movements within societies – like in India, as you well know, defining India as a Hindu country, though it’s not all Hindu.

I think we mentioned, in relation to Nigeria, the movement to call the northern part of Nigeria an Islamic region, instituting Shariah law. These movements are not necessarily supported by the central government, but they’re powerful movements within society.

SHAH: So you’ve got cases of failed states, like Somalia, where the government just isn’t effective at doing anything, let alone regulating religion.

GRIM: Yeah, that’s right. Each of these countries is a case study in itself to look at, but in terms of the trends, these are at least some that we’ve identified.

As political scientists, this could be coming from a variety of sources – the type of government; colonial history; mix of religious groups within the country, including religious groups, ethnic groups, etc. It could be coming from a variety of directions. Bill, this is actually my greatest concern in terms of how it’s going to be received and how it’s going to be reported on – that people will look at those regions, and it doesn’t take a genius to begin to assign certain religious tags to that. I can just guarantee you that that’s the way many of these reports out in the media are going to play out. It’s probably the thing we struggle with the most.

Do you hit it head on and then open a huge can of worms? Then we have to explain how you’ve got countries from religion X that rate very high on restrictions and countries with the same religion that rate very low. And this is true for every religious tradition. But once you raise that, you’ve got to then explain what accounts for the differences in country A versus B if they’re both of the same religion, both predominantly of this or that. I think that would have taken us way beyond the report.

But the risk that we run – and we understand it – is that people will very quickly look at how this thing maps out in the world, and they will draw certain conclusions in terms of what might be driving what we’re reporting on here. We, in good social science terms, remain completely agnostic on that question until we have done a lot more analysis on it. So it’s not a PC thing driving it for us; it’s really a commitment to good social science.

MARSHALL: Is this a very small religious group? (Laughter.)

UNIDENTIFIED: And they don’t like the homoscedastics at all! (Laughter.)

COOPERMAN: In the chart, this is pretty much all the 198 countries, not labeled but just as dots. You can easily see a couple of patterns jump out. One is that really, in fact, most countries are relatively low. They’re in the low or moderate range on most things. And then, generally speaking, by the fact that there are virtually no dots in the upper left-hand corner and very few dots in the lower right-hand corner, you see that they generally do follow.

So social hostilities do tend to move in tandem. It’s relatively rare to have a country that has high restrictions on one index but does not have high restrictions on the other. However, it’s not a tight correlation. There are a lot of exceptions, and we find that the exceptions are kind of interesting.

To get back to Paul Marshall’s initial observation about what’s surprising and not surprising in this report, we certainly hope – honestly – that not everything in here will accord with your prior views on things and that there will be some surprises. Maybe one of the surprises here is that some of the countries are kind of outliers – high on one and low on the other by our scores – and to ask ourselves the question of why that is and what the ramifications of that are, which may go beyond this report.

But in addition to this report, Pew is doing a lot of other work, and we are doing attitudinal surveys around the world. I suppose Luis would not like me to say very much in detail about it. But, for example, we have a huge survey in sub-Saharan Africa, and the results showing levels of social tolerance between religious groups are really quite interesting. They do give us confidence that what we’re finding, for example, in lower levels of restrictions in sub-Saharan Africa than in Europe accord with what we’re finding in our social surveys as well. We hope we’ll be able to share all that with you before too much longer.

LUGO: But if the results of the report, Alan, do correspond to everything you knew, keep in mind the famous saying of a former secretary of education, who was actually a philosopher by training, who once described the social sciences as the elaborate demonstration of the obvious by methods that are obscure. (Laughter.) So keep that in mind.

I have three questions. Perhaps in order of importance, I think the first one is related in some way to what we’ve been talking about here, and Brian and Alan and I have talked a little bit about this already, and that is the question of government restrictions on political activity by religious individuals and communities.

If you think of religious liberty as including the right, as I do and I suspect most do, the right of religious individuals, actors and communities to be involved in the political life of a community, and the restrictions that may exist on that, primarily governmental restrictions, do you feel that you’ve gotten at that in your questions?

I know that on page 61, question 20.2 talks about receiving the same level of government access and privileges, which you had mentioned, Alan. But to me, that doesn’t quite get to the kind of thing that I have in mind. For example, the French restriction, not on headscarves so much, but on bringing religious matters into the public square.

If you would address that, and then two smaller questions on the United States: Did you see in judging the restrictions on religious freedom the Establishment Clause as a qualification on religious freedom, broadly construed? And secondly, is the United States terrorism list viewed as a ban on religion – Hamas, for example? Is that a ban on a religious group, and therefore does that raise the U.S. score?

LUGO: All right. Several good questions in there. So Brian, let’s tackle the first, government restrictions on political activities of religious groups. How is that accounted for in the report?

GRIM: Tom, from your perspective, religious freedom implies or includes the right of religious groups to have a voice in public life, to be able to make their arguments in the public square about how the country should be run and organized.

We don’t have a direct question on that. Some of the questions get at that. As we mentioned before, the public preaching question – are they restricted in what they can say? So if they were wanting to preach and say, our country should be heading in this direction and we’re advocating for certain policies, and the government restricts them from making those statements, we would pick it up in that question. The same in the broadcasting questions. We would pick it up, but we haven’t addressed that directly.

Part of it is this coding instrument itself; these questions were developed by looking at what the sources give information on. Going back to my days at Penn State, I just read the State Department reports and then started going through and saying, well, here, they talk about that, I’ll write a question on that; they talk about this, I’ll write a question on that. Then as we’ve expanded it to look at other sources, we used those same questions. The State Department reports put together under your watch, back in the day, do not address this.

LUGO: Oh, that’s mean, Brian. Come on! (Laughter.) You’re a nice guy; you’re not supposed to –

GRIM: It’s Tom’s fault.

LUGO: It’s Tom’s fault! No, but even though there’s no direct measure, what you’re saying, Brian, is there are several proxy measures that get at that same issue.

GRIM: There are. Right. Yeah, and, one, the Establishment Clause – that wasn’t viewed as a qualification or a contradiction in our coding. And then regarding bans on religion, the Treasury Department more or less shut down the Holy Land Foundation based in Texas. Had our sources recorded that closure in their reporting, we may have counted that. However, just being on a terrorist watch list, that wasn’t counted as a ban. The reports we looked at didn’t say the group was banned in the United States.

So no, in the time period we were looking at, it wasn’t counted, though the Holy Land Foundation was closed down. One of the differences in the United States coverage is that we don’t have an international religious freedom report where maybe some of these things would be teased out a little bit more.

LUGO: A lot of the sources on social hostilities were from the Justice Department, right?

GRIM: Justice Department, yeah. Human Rights First report.

LUGO: How did we pick that up, Brian, in terms of government policies that incentivize social hostility in one way or another?

GRIM: I don’t know that we wrote a question to try to get at that, but many of the questions deal with whether or not governments use force in carrying out their policies and restrictions on religion. I think when a government’s used force against a religion, that would be an incentive for them to fight back. And that often is the case. As our correlation shows, where one side pushes, the other side tends to push back.

COOPERMAN: I immediately thought of two measures, and I’m trying to find them. One asked the question about whether the government did not intervene – ah, yeah. It’s question 13: “Were there instances when the national government did not intervene in cases of discrimination or abuses against religious groups?” So it wouldn’t get to whether the government incentivized it but whether the government essentially turned its back on it, allowed it to happen.

And there’s another question there – there was one other that I thought of. But suffice it to say, you’re right. There’s no direct question on whether the government is behind particular groups.

MCDONNELL: I was thinking of something like the government of Sudan, where they’re very good at using proxies to carry out the things that they want to do against religious groups.

GRIM: Yeah, if it’s a semi-governmental force or if it’s a force sponsored in some way by a government, we would count that as government action.

COOPERMAN: I suppose it’s worth pointing out – and we say this a number of times in the report – that the measures of government favoritism are carefully constructed, almost like an index within an index-in-waiting. It’s actually five questions rolled into one question. It looks unusually complex and it is. The reason for that is that we don’t consider it a restriction if a government supports religious groups through funding or other manners if it does so in a nondiscriminatory, neutral, way – if our sources don’t call it out as doing so in a discriminatory way. And so we ask a series of questions that are aimed at getting to government favoritism. But if it’s neutral, the government doesn’t get a negative score for it.

LUGO: This gives me the opportunity, Faith, to remind everyone that what Alan said at the beginning was not a throwaway line. As you read carefully through this report, if you have suggestions that would help us improve this product as we go forth – we do want to shorten the time from the coding that we do to the time of release and get a lot better. So the plan for the next round, for instance, is to begin January with all the 2009 information and get that out fairly quickly, by the middle of the year.

GRIM: Actually, tomorrow – there’s no rest – tomorrow, I begin.

LUGO: Ah, with the Georgetown folks, the Berkley Center. Well, that’s terrific. That’s very, very good. Again, thank you so much, folks. We appreciate you coming and participating, and we’ll stay in touch. Thanks. (Applause.)