Home » Library » Modern Library » Michael Martin Friendly

Michael Martin Friendly

Friendly Atheism (1996)

Michael Martin


Should atheists be friendly towards theists? Is what might be called friendly atheism to be encouraged? This question is not only interesting it is own right but has obviously practical implications for atheists in dealing with theists in our society. In this paper I will explore this question relating it to the connection between atheism and humanism as well as a recent argument advocating friendly atheism.

Humanistic Values and Humanitarianism

Atheism is often closely associated with humanism. It might seem off hand that the trait of being friendly must be associated with humanistic values. So atheism and being friendly should go together. One might look at it this way. Humanists, in one sense of that ambiguous term, are by definition concerned with human welfare. Thus, it might be argued that humanism is closely related to humanitarianism. Although what exactly humanitarianism involves is a complex issue, it might be said that it is plausible to suppose one aspect of the complex of traits that make-up humanitarianism is friendliness. Thus, one might suppose that a humanitarian would be friendly. But since humanism is closely associated with humanitarian one would suppose that a person with humanistic values would be friendly. But since atheism is associated with humanism atheists would be friendly.

However, I think this way of viewing the situation is mistaken. The connection between atheism and humanistic values is not as close as has usually been supposed. One can be an atheist and not advocate humanistic values and one can be a theist and support such values.[1] But even if I am mistaken on this point, there is not a close connection between humanistic values and humanitarianism and not a close connection between humanitarianism and friendliness.

What is a humanistic value theory? Although the specific values presupposed may differ from one humanist to another, all humanists believe that moral values are relative to human experience. As one commentator has put it, humans “are the measure of all things.”[2] What does this mean? One plausible interpretation of the thesis that humans are the measure of all things is that all intrinsic moral values are based on human desires and interests. Other things could have value but they would have only instrumental value; that is, they would be means to furthering human desires and interest. For example, a humanistic value theorist might maintain that a work of art has instrumental value in furthering human aesthetic enjoyment but no value in its own right. Such a theorist might also argue that human aesthetic enjoyment has intrinsic value; that is, it has value in its own right.

There is another interpretation of the claim that humans are the measure of all things, however; namely that humans rather than some divine being decide what is morally valuable. On this interpretation, human beings, not God, create values; in other words, values are made by human decisions not discovered. Of course, one could give different accounts of what values humans have created. On this interpretation, moreover, works of art could have intrinsic, not just instrumental value.

It seems clear that humanistic value theory in the second sense has nothing to do with humanitarianism. Just because values are made by human decision does not mean that these values have anything to do with human welfare. Indeed, human value theory in this sense is compatible with human interests and desires having neither intrinsic nor instrumental value. But what about the first sense of humanistic value theory? Although it is less obvious this sense also has nothing to do with humanitarianism in its usual sense. Humanitarianism is usually closely associated with charity and philanthropy and in general with an altruistic promotion of human happiness and well being. However, one might suppose all intrinsic moral values are based on human desires and interests and yet not suppose that there is any value in charity and philanthropy and in general in any altruistic promotion of human happiness. Indeed, a believer in humanistic values in the first sense might suppose that charity is a bad thing, and altruism is a vice. Humanist value theory in this sense is perfectly compatible with a rugged individualism and a militant egoism that eschews humanitarianism of any kind.

Humanism, Humanitarianism and Friendliness

Although humanistic value theory does not entail humanitarianism one might well suppose that humanitarianism does entail friendliness. But the relation between humanitarianism and friendliness is tenuous. Friendliness usually connotes openness, warmth, and sociability in personal relations. Yet humanitarianism does not seem to have any such implication. A person could be warm, open and social and not be a humanitarian. For example, a politician might be friendly and yet not be known for his humanitarianism. Conversely, a great humanitarian could be cold, secretive, and unsociable in her personal relations.

Is there no connection between friendliness and humanitarianism? Does not to say that X is friendly toward Y normally imply that X is willing to help Y in case of need? However, to say that X is a humanitarian normally implies that X is willing to help Y in case of need. So it would seem both ideas imply being willing to help in case of need. However, one must not confuse being friendly toward someone and being a friend of someone. Being a friend towards a person does normally imply being willing to help this person. But it is unclear if being friendly towards someone has this implications. It would seem that X’s being warm, open and sociable towards Y is quite compatible with X believing that people should look after themselves and that humanitarian aid is to be avoided. Even if there is this similarity, there are crucial differences. To speak of X as friendly towards Y, suggests that X is dealing with Y on a personal level. But to speak of X as rendering humanitarian assistance to Y does not imply this. Indeed, the attribution of humanitarian concern to someone is compatible with an impersonality and a distance that is incompatible with being friendly.

If there is no clear relation between friendliness and humanitarianism, there is even less of a clear relation between humanistic theory of value and friendliness. In the first sense humanistic value, that is, that all intrinsic moral values are based on human desires and interests, nothing follows about the value of friendliness. The irrelevance of humanism in the second sense is even clearer. If humans create values, nothing follows about the value of friendliness.

Atheism and Friendliness: Rowe’s Analysis

It is obvious that there is no conceptual connection between atheism and friendliness. As far as the logic of the concepts are concerned one can be either a friendly or unfriendly atheist or a friendly or unfriendly theist. But this does not settle the normative question. Historically atheists have been persecuted and discriminated against and even today atheists often live in a hostile atmosphere. Given this background should atheists be friendly? According to the well known philosopher of religion, William Rowe [3] atheism should be friendly. But is Rowe correct?

According to Rowe an unfriendly atheist [4] believes that no one is justified in believing that the theistic God exists; an indifferent atheist holds no belief concerning whether any theist is or is not rationally justified in believing that the theistic God exists; a friendly atheist believes that some theists are rationally justified in believing the theistic God exists.

Rowe defends the position of friendly atheism. First, he argues that one can be justified in believing a false proposition. Thus, atheism may be true and yet a person can be justified in believing that it is true that the theistic God exists. Second, he argues that it is possible for an atheist to be justified in believing that the theistic God does not exist and for a theist to justified in believing this God exists. How can this be?

According to Rowe it is quite clear how this could be the case if one is talking about what an atheist is justified in believing now and what, for example, some theist in the eleventh or thirteenth century were justified in believing. Given our present evidence the modern atheist would be justified in believing that the theistic God does not exist relative to this evidence while a theist living the Middle Ages would be justified relative to the evidence available then, in believing that the theistic God did exist. However, Rowe maintains:

The more interesting question is whether some people in modern society, people who are aware of the usual grounds for belief and disbelief and are acquainted to some degree with modern science, are yet rationally justified in accepting theism. Friendly atheism is a significant position only if it answers this question in the affirmative.[5]

Rowe goes on to argue that:

It is not difficult for an atheist to be friendly when he has reason to believe that the theist could not reasonable be expected to be acquainted with the grounds for disbelief that he (the atheist) possess…..Friendly atheism becomes paradoxical, however, when the atheist contemplates believing that the theist has all the grounds for atheism that he, the atheist has, and yet is rationally justified in maintaining his theistic belief. But even so excessively friendly a view as this can perhaps be held by the atheist if he also has some reason to think that the grounds for theism are not so telling as the theist is justified in taking them to be.[6]

Rowe presents in a footnote the following example to illustrate the possibility that he alludes to in the last sentence of the above quotation.

Suppose that I add a long sum of numbers three times and get result x. I inform you of this so that you have pretty much the same evidence that I have for the claim that the sum of the numbers is x. You then use a calculator twice over and arrive at result y. You, then, are justified in believing that the sum of the numbers is not x. However, knowing that your calculator has been damaged and is therefore unreliable, and that you have no reason to think that it is damaged, I may reasonable believe not only that the sum of the numbers is x, but also that you are justified in believing that the sum is not x. Here is a case, then, where you have all of my evidence for p, and yet I can reasonable believe that you are justified in believing not-p — for I have reason to believe that your grounds for not-p are not as telling as you are justified in taking them to be.[7]

Evaluation of Rowe’s Analysis

Unfortunately, although Rowe gives this example to illustrate what he means, he does not give any examples that have to do directly with friendly atheism. Even if the example showed what Rowe thought it does,[8] it is by no means obvious that an analogous example can be found that applies directly to friendly atheism. Indeed, it is very difficult to see how in any realistic way present confrontations between atheists and theists can be represented by this example. In Rowe’s math example one person does not have access to certain crucial information, that is, that the calculator is damaged – information that is available to the other person. Indeed, the person with the information could immediately show the other person that he or she is not justified in believing that the sum is y by revealing this information.

But in the modern controversies between atheists and theists atheists are not holding back information. According to Rowe theists may attempt to justify their belief by appealing to one or more of the traditional arguments for the existence of God or by appealing to certain aspects of religious experience or by arguing that theism is "a plausible theory in terms of which we can account for a variety of phenomena."[9] Atheists’ critiques of these three theistic attempts to justify their beliefs should be well known. If theists do not know these criticisms, it is surely not because they are inaccessible. Because they are so widely accessible it can hardly be claimed that theists are justified if they do not take them in to account. Thus, atheists may believe that the theistic arguments are not as telling as theists suppose they are but it is hard to see how atheists could believe that theists are justified in holding that these arguments are as telling as theists suppose they are. For atheistic literature consists largely in arguing that theistic arguments are as not as telling as theists suppose and that atheistic arguments are more telling than theists suppose and that theists have done little to show otherwise. The only situation that is at all analogous to Rowe’s example is where a theist may been deprived of access to atheistic arguments and literature. But this does not seem to the sort of situation that Rowe has in mind when he speaks of friendly atheism in the context of the theist "being acquainted with the grounds for disbelief that he (the atheist) possesses."[10]

I conclude that friendly atheism as Rowe conceives it should not be defended since in terms of the modern confrontation between atheism and theism it is not a plausible, or perhaps not even a coherent, position.

Friendly Atheism Reconsidered

As Rowe conceives of a friendly atheist he or she is a person that believes that some theists are rationally justified in believing a theistic God exists. It seems that this definition has little to do with the ordinary meaning of ‘friendly’. According to the dictionary ‘friendly’ is defined as having the disposition of being open, warm, and social. In the ordinary sense of ‘friendly’ one can be friendly toward people who one believes holds rationally unjustified beliefs and one can be unfriendly toward people that one believes holds justified beliefs. Furthermore, despite this a friendly or unfriendly attitude may be justified. For example, one can be friendly towards a person in mental institutions who one believe holds a completely wrong view of the world, e.g, he thinks he is Jesus Christ and who has no rational justification for his beliefs. One may be friendly towards this person because he is likeable or has been kind or good to you. The irrationality of his beliefs does not enter into your motivation of your friendly behavior.

Perhaps, then, one should consider friendly atheism in a way that is more in accord with the ordinary use of the term ‘friendly’. To advocate friendly atheism in this sense of the term ‘friendly’ would be to advocate that atheists be friendly people, that is, that they should be open, warm and sociable. This would include being friendly to theists, their traditional enemies. Is such a position plausible?

There would seem to be no good reason why an atheist could not claim that a theist is wrong and unjustified and yet be friendly toward the theist. Indeed, there may be good reasons for such an attitude.

It is possible such friendly behavior would help to eliminate a commonly held stereotype of the atheist as an embittered, suspicious loner snarling against the theistic culture he or she is forced to live in. (Whether there is any evidence that supports such a stereotype I do not know.) So more friendly atheists may be a good thing as a way of improving the social image of atheism.

Furthermore, in so far as atheists want theists to ‘convert’ to atheism, a friendly attitude is no doubt desirable. Just as religious missionaries may improve their success rate by a more friendly and less judgmental attitude to potential converts, so atheists ‘missionaries’ would possibly have a higher success rate if they adopted a less confrontational and more friendly attitude.

However, in these justifications of friendly atheism the friendly atheist may appear disingenuous using his or her friendly behavior for ulterior purposes. There is this danger, of course, but one might consider these uses of a friendly attitude as side effects of a genuine friendly attitude that would be fostered for other reasons. What are these other reasons? First of all, an atheist may be friendly towards theists because he or she likes them and believes that they need help. He or she may believe that they are deeply confused people who are living a lie. The atheist may genuinely wish to help theists to see the truth and change their lives. Friendly atheism may, then, be an outpouring of compassion. In addition and less altruistically, the atheist may believe that he or she has much to learn from theists. There is no inconsistency in an atheist believing that many things that theists believe are interesting, insightful and even profound and believing that their position on the existence of God is wrong and unjustified. Finally, atheists may greatly admire what some theists do. There is no inconsistency in an atheist believing that some of the ethical beliefs that are divorced from any theological underpinning are true and that some practices of theists are important and worthwhile. Because of this admiration friendship may grow.

But are there good reasons for atheists not to be friendly towards theists? Although the days of the Inquisition and wide scale religious persecution are over, at least in Western democracies some religious zealots in the United States, for example, are attempting to legally imposing their views of religious morality.[11] No doubt atheists as well as others should attempt to fight such encroachment. But what place does a friendly attitude have in such difficult controversies? Can and should atheists be friendly to those theists who seek to deprive them of their religious freedom and bridge the separation of Church and State? Perhaps the best strategy for the atheist to adopt is an experimental attitude. The atheist should attempt to be friendly in so far as this persuades the misguided zealots from their course and yet be ready to adopt other attitudes if being friendly does not work.


I conclude that friendly atheism as I have constructed it is a plausible view and atheists would be well advised to give it serious consideration. However, I have shown that this judgment cannot be supported by associating atheism with humanism or humanitarianism since there is no clear connection between these three "isms" and no clear connection between them and friendliness. Moreover, I have rejected William Rowe’s argument for friendly atheism as both unsound and misleading. If friendly atheism is to be justified, it must be pragmatically in the context of atheistic-theistic relations.


[1] See my paper Michael Martin, "Atheism and Humanistic Value Theory," Sept. 26 /library/modern/michael_martin/values.html

[2] Nicola Abbagnano, "Humanism," Encyclopedia of Philosophy, Vol. 4, (New York: Macmillan Publishing Co. Inc. and the Free Press, 1967) p. 72.

[3] William L. Rowe, "The Problem of Evil and Some Varieties of Atheism", American Philosophical Quarterly, 16, 1979, pp. 335 – 341. See also William L. Rowe, Philosophy of Religion (Belmont,Calif. : Wadsworth Publishing Co.1978), pp. 93-94.

[4] Rowe defines a broad and narrow sense of ‘atheist’ and ‘theist’. In the narrow sense a theist is someone who believes in the existence of omnipotent, omniscient, eternal , supremely good being who created the world. An atheist in the narrow sense is a person who denies the existence of such a God. A theist in the broad sense is someone who believes in the existence of some sort of divine being or reality. An atheist in the broad sense is someone who denies the existence of any sort of divine being or reality. Following Rowe in the paper I will be using ‘theism’ and ‘atheism’ in the narrow sense. However, much of what I say is applicable to ‘theism’ and ‘atheism’ in the broad sense.

[5] Rowe, “The Problem of Evil and Some Varieties of Atheism”, p. 340.

[6] Ibid.

[7] Ibid., pp. 340 -341, n. 7.

[8] Rowe’s example is surely misleading. It does not show what he supposes. The person who does the sum with the damaged calculator does not have all the relevant information that the other person has. This person lacks the information that the calculator is damaged.

[9] Ibid., p. 340.

[10] Ibid., p.340.

[11] Another case where a nonfriendly attitude may be justified is when theists attack an atheist’s character. I have experienced such an attack and I believe an unfriendly attitude towards those who attacked me is justified. See Michael Martin, “Reply to Bulter, Ventrella, and Field,” Sept. 3, 1996 https://infidels.org/library/modern/michael_martin/reply.html

"Friendly Atheism?" is copyright © 1996 by Michael Martin. All rights reserved.

The electronic version is copyright © 1997 by Internet Infidels with the written permission of Michael Martin. All rights reserved.

all rights reserved