On the Compatibility of Christian Faith and Theological Agnosticism

: Agnosticism is often understood as being in opposition to religious faith. This article ex-amines the two concepts “agnosticism” and “Christian faith” and their interrelated character. We argue that agnosticism and Christian faith are compatible, although agnosticism has some negative consequences for Christian faith seen from a Christian perspective.


Introduction
In public discourse, the terms "Christian" and "agnostic" are often used as incompatible alternatives. Either you are a Christian or you are an agnostic-you cannot be both. Atheism and agnosticism, on the other hand, are often conflated so that people are not sure whether they are atheists or agnostics but, and that is often the decisive part, they do not go to church. A few years ago, in a public debate, the spokesman of the Danish atheist society ("Ateistisk Selskab"), the journalist Anders Stjernholm, admitted that he was not an atheist but an agnostic, that is, he claimed that neither he nor anyone else knows whether or not God exists (Konnekt 2016). This admission did not make him quit the atheist society. Instead, he later became their chairman. Bertrand Russell famously had similar troubles. In a short text from 1947 titled "Am I An Atheist Or An Agnostic?" he stated the following: "I never know whether I should say 'Agnostic' or whether I should say 'Atheist'. It is a very difficult question and I daresay that some of you have been troubled by it. As a philosopher, if I were speaking to a purely philosophic audience I should say that I ought to describe myself as an Agnostic, because I do not think that there is a conclusive argument by which one prove that there is not a God. On the other hand, if I am to convey the right impression to the ordinary man in the street I think I ought to say that I am an Atheist." (Russell 1947).
This separation between Christians on the one hand and atheists and agnostics on the other is often sustained by the discourse in Christian communities. Christians will identify with the term "Christian" and thereby reject to be identified by the terms "agnostic" or "atheist" since they are "believers." However, this identification does not necessarily mean, that they are not agnostics in a technical sense. If instead of asking someone who identifies as a Christian whether or not she is agnostic, you ask her whether she has any evidence to support her Christian faith, she will often admit that she does not have any evidence to support her faith, which is exactly what agnostics often claim. This reveals a muddled relationship between Christian faith and agnosticism in public discourse.
In this article, we will seek to clarify the relation between theological agnosticism and Christian faith by assessing whether someone with Christian faith can also be an agnostic. called the traditional or standard view of beliefs, which we will assume in this article (for a presentation and discussion of alternative views of belief, see Heil 1992a, pp. 45-46;Schwitzgebel 2019;Moser 1995, pp. 78-79). When someone believes that a certain car is blue, he will have the psychological state of holding the proposition "the car is blue" to be true. Now, what creates or determines our beliefs? In ordinary discourse, we often talk as if we chose our beliefs. On reflection, this seems unlikely. If I see a closed door in front of me, I cannot choose to believe that the door is open. I can choose to act as if it is open (thereby banging my head into it), but I cannot bring myself to believe it is open. This example reveals the important distinction between acting on a belief and acting on an assumption (which is often muddled in ordinary discourse) (Swinburne 2005, p. 4). Even though I cannot choose my beliefs, I can still choose to act on the truth of a proposition which I do not believe.
So, if I do not choose my beliefs, then how do I acquire them? According to John Heil, "acquiring a belief is like catching a cold" (Heil 1992a, p. 47). It is not something you choose. It is something that happens to us due to external exposure. Richard Swinburne likewise emphasizes the external factor in belief-formation. He states, "we believe our beliefs because we know that we do not choose them but because (if we think about it) we believe that they are forced upon us by the outside world" (Swinburne 2001, p. 40). If this is true, then what is it that forms our beliefs? This brings us to the important concept of evidence. Evidence is the external factor that shapes our beliefs (for a discussion of belief and the nature of evidence see Price 1969, pp. 92-129). By evidence, we understand what Francis Jonbäck names as "truth indicators," that is, "something that indicates the truth of a proposition 'p' for a person 'S'" (Jonbäck 2017, p. 11). Here, we must distinguish between what we name subjective and objective evidence. Subjective evidence is what "S" regards as truth-indicating (that is, what actually forms her beliefs) while objective evidence is what actually is truth-indicating (that is, what should form her beliefs if she wants true beliefs). In this article, we have chosen to use evidence in a subjective sense for two reasons. First, the aim of this paper is not to assess whether an ideal rational agnostic can have Christian faith but whether real agnostics can have Christian faith. Secondly, as we shall find below, belief is a central element of Christian faith, but having rational beliefs is not a requirement for having Christian faith. Some Christians might have rational faith, others irrational faith, but what interests us here is that they have faith.
Evidence is something other than proof. To have evidence for a belief is not to have a proof for it but only that something counts as indicating that "p" is true. In probabilistic terms, a proposition is proved if its probability is 1. A proof is an argument which confers the probability of 1 upon a proposition. Evidence comes in several kinds. Sense perception is one kind of evidence (I believe the car is blue because I can see it is blue.). Social evidence (or testimony) is a second kind (I believe the car is blue because John has told me it is.). Intuition is a third kind (I believe "2 + 2 = 4" because it seems self-evidently true.) More could be added. It is important to note that evidence is not an infallible category. Evidence always exists in large philosophical frameworks that determine whether a piece of data has evidential weight in relation to a certain proposition (I have no reason to believe the car is blue if I assume my sense-data is a computer simulation in the Matrix). It is often assumed, that either you believe a proposition, or you disbelieve it, or you are agnostic about it. But beliefs are not either-or. According to David Hume, "a wise man […] proportions his belief to the evidence (Hume 1988, p. 101)." Some beliefs we hold strongly, others weakly. Or we might say that some statements have a high/low degree of credence/subjective probability/conviction for us. This is because we do not have the same amount of evidence for all the propositions we believe (Price 1969, p. 39). For example, even though we cannot prove with absolute certainty that other persons exist outside our own mind (that is, disprove solipsism) most will find the amount of evidence in favor to be overwhelming, and thus find themselves with a strong belief in the existence of human persons outside their own mind. Here we can point to Bayesian probability theory that has been developed to express the logical relation between evidence and the truth-probability of a hypothesis. Richard Swinburne claims, that one believes a proposition or hypothesis when P(h) > 0.5, that is, when the probability of the hypothesis is above 0.5 compared to the negation of the hypothesis (P(¬h) < 0.5) (Swinburne 2001, pp. 35-36, pp. 102-107). This diverges from the way we ordinarily talk about beliefs. If someone finds that the evidence is slightly in favor of God, we might still find it more suitable to name him an agnostic than a believer. Despite this, Richard Swinburne still suggests that we use the term "belief" in this technical sense. For a discussion of belief and probability, see (Swinburne 2001, pp. 56-128).
The last aspect of belief we need to note is the fact that we can change and improve our beliefs over time. We have stated above that we cannot change our beliefs by will, only our actions. However, since our beliefs are formed by the evidence we have; we can try to change our beliefs by seeking out new evidence. A person who, for example, encounters the cosmological argument for the existence of God might upon first acquaintance come to believe in God but, upon further investigation, come to find flaws in the argument making the argument lose its initial evidential weight.
This short exposition of the concept of belief gives us a framework for defining and understanding agnosticism. The last thing we need to clarify is what we mean by theological agnosticism.

Defining Theological Agnosticism
Having now examined the doxastic aspect of agnosticism, we can now proceed to comment more specifically on the kind of agnosticism which we will assess in this article. So, what is agnosticism? We define agnosticism as a doxastic attitude of subject S towards a proposition p where S does neither believe nor disbelieve its truth. In Bayesian terms, S believes that P(h) = 0.5. Since such a position is (infinitely) narrow, it might make more sense to define agnosticism as a probability interval (like 0.4-0.6). Whether we define it one way or another does not impact (much) on the rest of the article. We are, of course, not just interested in any kind of agnosticism but only in agnosticism about the existence of the Christian God. By "theological agnosticism," we therefore mean that S does neither believe nor disbelieve in the existence of the Christian God. Since belief is a doxastic attitude toward a proposition, theological agnosticism needs to be related to a statement (or a network of statements) about what is meant by "the Christian God." Such an analysis is encompassing and unnecessary at this point. We will just note that the understanding of the Christian God which we, the authors, have in mind, is something along the line of the apostolic tradition (expressed for example in the Apostolic Creed). We assume that we can speak meaningfully about God (and therefore that our sentences about God can be true or false) but not that our language in any way can fully capture the nature of God. There is, of course, discussion among theologians about whether and how we can describe God. Since belief is a doxastic attitude towards a proposition, we need to assume that God can be described by human language-at least to some degree. This is controversial among some theologians. Non-cognitivist theologies claim that theological sentences like "God is love" does not have a truth-value but might function to express emotions (emotivism) or regulate behavior in a community (imperativism). This makes it impossible to believe or disbelieve theological sentences. We need not settle this dispute here. We just need to note that the arguments in this article assume theological cognitivism.
Having now defined the concept of theological agnosticism, we need to turn to the question of the nature of Christian faith.

What Is Christian Faith?
Before we engage in identifying the nature of Christian faith, we need to distinguish between two concepts of faith. The first concept of faith is the ordinary loose use of the term to denote adherence to a religion, worldview, or ideology. The second concept of faith is the theological concept found in the Christian religion. The aim of this article is to examine whether the latter is compatible with agnosticism. Therefore, we are not just interested in defining a general religious faith, instead we want to define the contours of a particularly Christian conception of faith.
So, what is Christian faith (in the second sense)? This is, of course, a controversial question among Christians. We cannot, therefore, answer our questions about the compatibility between agnosticism and Christian faith on behalf of Christianity as a whole. We have therefore chosen to give our own definition of faith by drawing inspiration from a number of Christian thinkers.

On the Nature of Faith
The Christian philosophers J. P. Moreland and W. L. Craig use the three terms notitia, assensus, and fiducia to explicate the nature of faith. It is often claimed that the three terms are derived from Loci Communes (1521) by Melanchthon. But Melanchthon does not use the term "notitia" in Loci (although he uses the term "opinio" for something similar) and he does not distinguish as systematically between them as Moreland and Craig does (cf. Loci Communes "De Justificatione et Fide" pp. 197-220 in Melanchthon 1864). Notitia is the content of faith. This is parallel to the role propositions have in how we defined belief above. Assensus is the person's adherence to the truth of the proposition. This, again, seems parallel to how we defined belief above as a doxastic attitude of a person-although the term "adherence" carries connotations that faith is a choice. Last, and most important, faith is fiducia, trust in God (Moreland and Craig 2003, p. 18). The element of trust adds something to faith that is not contained in the former two. In the philosophical discourse on belief, philosophers often distinguish between "belief-that" and "belief-in" (fides qua/fides quae). Belief-that is propositional belief while belief-in is belief in a person, that is, trust in a person. Belief-that is the "faith" of notitia and assensus, while belief-in is the faith of fiducia, which presupposes (or contains) notitia and assensus. It has often been discussed whether belief-in (trust) can be reduced to belief-that (propositional belief). Aquinas claimed that to believe in God is simply to believe in the truth of certain statements. H. H. Price has argued against this claiming that although a belief-that analysis of what it is to believe in someone captures many aspects of it, it does not seem to be able to catch its emotional component (Heil 1992b, p. 49, see also Swinburne 2005, pp. 142-47 andPrice 1969, pp. 426-54). Faith, therefore, seems to be more than just a doxastic attitude (a belief). It contains an emotional component, which is distinct from belief. Martin Luther makes a similar distinction between types of faith in a lecture from 1535. He states, that "the true faith embraces with outstretched arms and joy the Son of God given for it and say: 'My beloved is mine and I am his.' […] It is this 'for me' or 'for us' which creates the true faith when it is believed and which separates it from all other kinds of faith which only hears about the events (Luther 1926, W.A.39.I,46, own translation from, "Fides vera extensis brachiis amplectitur laeta filium Dei pro sese traditum et dicit: Dilectus meus mihi et ego illi […] Igitur illud, pro Me, seu pro Nobis, si creditur, facit istam veram fidem et secernit ab omni alia fide, quae res tantum gestas audit")." The true faith (fides vera), that is, Christian faith, is separated from the acquired faith (fides acquisita) by its emotional component. The fact that the gospel of Jesus concerns the believer is a source of joy, love, and faith. Allow a short excursus: According to Martha Nussbaum emotions are egoistic in nature. They concern something that is important to the one who has the emotions (Nussbaum 2003, pp. 30-31). This explains why "pro Me" makes such a big difference for Luther. It is only when the believer believes herself to be a part of the Christian gospel, that the gospel becomes an emotional reality. Only this kind of faith is justifying faith, according to Luther (Luther 1926, W.A.39.I,46,). The British theologian Alister McGrath (b. 1953) offers a similar three-factor analysis of Christian faith, which adds a fourth component to the three described above. The first aspect of faith, according to McGrath, is the doxastic aspect, that is, "faith is about believing that certain things are true (McGrath 1993, p. 49, italics removed)." And what are these "things"? It is mainly "belief in the existence of God and his promises (ibid. p. 49)." His first factor thereby covers what is meant by notitia and assensus (or belief-that) above. The second aspect of faith is trust. To believe in God's promises is more than to assent to their propositional content. "It is an awareness that they can be trusted and relied on (ibid. pp. 49-50)." Faith is something that touches "the heart," something that is "a response of our whole person to the person of God (ibid. p. 50)." Finally, "faith is entry into the promises of God, receiving what they have to offer (ibid. p. 50, italics removed)." This last aspect is something that requires response or action on part of the believer. He states, "I may believe that God is promising me forgiveness of sins; I may trust that promise; but unless I respond to that promise, I shall not obtain forgiveness" (ibid. p. 50). The effect of faith is that "[it] unites us with the risen Christ and makes available to us everything that he gained through his obedience and resurrection-such as forgiveness, grace, and eternal life" (ibid. p. 51). The last element, therefore, contains two aspects. First, a behavioral element. Faith is responding or acting in light of her belief and trust in God. Second, a soteriological consequence. Through faith, the believer gains salvation.

Defining a Five-Factor Model of Faith
Based on the above examination, we propose to understand the concept of Christian faith as having five elements. First, faith requires propositional content, that is, a statement of what Christian faith is in. By using the term "propositional" we do not imply an adherence to the existence of a universal, eternal language which perfectly fits reality. What we mean is that faith requires a verbal expression of what it is directed at. Such a verbal expression is never perfect, but some expressions fit better than others with how reality actually is. This might be descriptive statements like "God exists" or promises like "anyone who comes to me I will never drive away" (John 6:37b NRSV). Of course, such statements are never isolated statements. Their meaning is derived from the network of statements which is implied by the context of the statement. A statement like "God exists" only makes sense because it is a part of a network of meanings (other statements) like "God is love" and "God is omnipotent" which are implied by the first. These statements (or propositions) have truth-value, that is, they can be true or false. If they did not, we could not form beliefs about them. We might call this aspect of faith, the aspect of "Christian theory" or simply "doctrine." How a person relates himself or herself to this "theory" or "doctrine" is captured by the next four elements of faith.
The second aspect of faith is the doxastic aspect, that is, faith is belief in the truth of a propositional statement. To have Christian faith, a person must believe in the truth of certain statements about God's nature, character, promises, etc.
The third aspect of faith is evidence, since, as we have seen above, evidence plays the central role in our belief-formation. Faith requires evidence because evidence shapes our beliefs about the Christian God. This is a controversial claim from a theological perspective. Faith is a result of God's acting, not human reasoning, some will object. The problem with this objection is that its conception of evidence is too narrow. The category of evidence (as we understand it) is not only limited to philosophical arguments but also includes categories like the testimony of one's parents, the Bible or one's church, and religious experience. Above, we have defined evidence as subjective evidence, that is, as whatever S takes to be evidence (whatever forms his beliefs). Of course, not everything taken to be evidence is actually evidence (in an objective sense). If the concept of evidence is construed in a more objectivist sense, the objection might be sound. But in the context of this article it does not follow from the necessity of evidence argued here, that faith is at the mercy of philosophical arguments.
Fourth, faith has an emotional aspect (see Westphal 2012, pp. 174-75 for a discussion of the nature of emotions). Merold Westphal states that "emotions […] can be seen as complex wholes with three distinct aspects: caring, believing, and feeling" (Westphal 2012, p. 175). Emotions contain beliefs and caring about something or someone. Caring is a measure of what is important or valuable for us. We do not have emotions for things we do not care about. Although we might on reflection think our caring mistaken. A person might reflect that he cares too much about whether his new car gets a scratch and too little about starving children in poor countries. Feeling is the last element of emotion and it only states the obvious, that an emotion is an experience, something we feel. Christian faith is having certain emotions in relation to God. Christian faith having the emotion of trust in relation to God and his promises. Other emotions might also be a part of Christian faith like love of God or longing for God. Some will object to defining faith as partly having certain emotions that most Christians do not have the emotions of trust and love in relation to God most of the time (although they regularly practice their faith). However, this objection might be met by the consideration that trust and love are "silent" emotions. They become most powerful when their object is disturbed. When our trust is let down, we feel betrayed. When we lose someone, we love we are overcome with sorrow. These powerful emotions show that trust and love was there to begin with even though there was not much to feel before the disturbance. Like belief, we cannot choose our emotions directly by will. Emotions are a reaction to what we believe (and care) about the world (Clore 2012, p. 217). To come to trust God, we must discover a trustworthy God. Even though certain emotions may have intrinsic value, one of their main functions is to motivate behavior (Clore 2012, pp. 214-16). This brings us to the fifth element of faith. It has a behavioral aspect. An aspect of faith is to act in a certain way. Such actions might be visible acts like prayer, going to church, taking communion, or caring for the needy or invisible acts like "giving one's life to Jesus." When a person comes to trust God and God's promises, this trust will motivate him or her to act in ways that are consistent with this trust. It is sometimes discussed whether acts should be conceptualized as an essential part of faith or as something which follows from faith. From our perspective, this is primarily a theoretical discussion, since in actual life our emotions will motivate our behavior. Whether or not a person trusts God will (almost) inevitably find expression in actions, although not necessarily in visible actions. Now, what is Christian faith? From the above analysis, it seems clear that faith is a complex, multilayered concept. What we see is that faith is no single distinct entity. Christian faith is more like a certain complex pattern in the totality of a person's beliefs, emotions, and behavior. We might call it a certain kind of lifeform, something which has a recognizable nature, but which cannot be reduced to one distinct entity. Christian faith is believing and trusting the Christian God and acting in ways compatible with this belief and trust. As we have seen above, there is a causal (but not deterministic) link between the different aspects of faith. We believe in a certain set of statements because we find evidence for them. This belief awakens certain emotions, which motivates us to certain acts. This reveals that something will happen to faith when a Christian believer becomes an agnostic. But what happens? And is it destructive of faith? We shall turn to those questions now.

Problems with the Compatibility of Christian Faith and Theological Agnosticism
In this section, we will consider the compatibility of Christian faith and theological agnosticism. First, we will consider the three most obvious objections, namely that agnosticism is incompatible with having Christian belief, trusting God, and living a Christian life. Then we will consider whether distinguishing between doxastic and evidential theological agnosticism can aid the case for some kind of compatibility. Last, we will question a premise underlying the former discussion, which turns out to be an argument for a certain kind of compatibility.

Faith Requires Belief, the Agnostic Does Not Have Belief
The first problem states the obvious. Agnosticism is incompatible with Christian faith because faith requires belief, and this is exactly what the agnostic does not have. The problem arises because we, so far, have assumed that the aspects of faith are of equal importance and necessity. However, it has often been claimed that belief is secondary to trust and living Christianly. Usually, the three go together, but the importance lies with the latter two. Doubt (which is quite similar to the concept of agnosticism applied here) is often recognized as a common part of being a Christian. Doubt alone is not enough to disrupt faith. Something more is needed (see, for example, McGrath 2006). For McGrath unbelief (the opposite of faith) is not doubt, but "an act of will" (McGrath 2006, p. 13).
As is stated in The Letter of James, "You believe that God is one; you do well. Even the demons believe-and shudder" (Jam 2.19 NRSV). This "faith of the demons" is not identical to Christian faith, which is more than mere belief in God's existence. If it is assumed that trusting God and living a Christian life is the decisive aspects of faith, then agnosticism does not by definition rule out faith. The question then becomes whether a person can trust God and live a Christian life without positive belief. Whether this is possible, we will discuss in the next two sections. Note, however, that Paul Draper questions whether the "Leap of Faith" of an agnostic should be counted as real faith since it seems to involve confident belief which is an intellectual vice for the agnostic because he lacks evidence. If faith is a virtue it should not require a vice (Draper 2002, p. 206). Our Christian perspective presented here claims that the agnostic need not violate his epistemic duties in his "Leap of Faith" since belief is a non-essential part of Christian faith.

Trust in God Requires Belief in His Existence
In sermons, trust in God is sometimes explained through an analogy of the trust between a boy and his dad. The boy climbs a high fence and asks his father to catch him when he jumps down. Without doubting the boy jumps trusting that his father will catch him. Now, would the child trust his father to catch him, if he did not know whether his dad was there or not? Probably not. The analogy is, of course, limited in explaining the relationship between God and humans, but it catches an important gist. How can you trust God if you do not believe that he exists? This problem might be solved in two ways.
First, it is a traditional theological claim that love of and trust in God is only possible because of an act of the Holy Spirit. Christian faith is not something humans can choose or produce; it is a gift of God. One way of interpreting this doctrine is as the claim that God sometimes miraculously induces someone with a feeling of trust regarding himself. This seems to solve the problem, enabling the agnostic to trust God, but it seems questionable whether the agnostic remains an agnostic after such an intervention. The Holy Spirit may not only awaken certain emotions but also beliefs. But even if this is not the case, then the new emotions of trust and love might create belief. We have noted above that there is a psychological link between belief and emotions.
Our emotions are based on our beliefs about the world. However, the opposite is also true; our beliefs are influenced by our emotions (Clore 2012, p. 217) as See, for example, Haidt has showed in his discussion on moral judgments, reasoning, and intuition (Haidt 2008). Haidt claims that our moral judgments are (most often) not based on our moral reasoning but on moral intuitions (emotions). Our moral reasoning is, according to Haidt, a post hoc construction which we primarily construct to justify our beliefs to others. It is a two-way street. So, if the Holy Spirit awakens emotions of trust and love, belief will probably follow. This suggests that the agnostic will not stay an agnostic after an intervention of the Holy Spirit. So, this seems not to solve the problem.
Second, emotions can be awakened by our imagination. Ordinarily, our emotions depend on our beliefs. If, for example, John tells Peter how he recently lost his mother in an accident, Peter will feel sorry for him, but if John continues and tells Peter that the story of his mother was a lie and that she is actually well, Peter's feeling of compassion for John will disappear. While this makes perfect sense, the curious thing is that through the imagination we can come to feel real emotions for fictional characters knowing they are fictitious (Radford et al. 1975;Gendler and Kovakovich 2010;Bloom 2011). We can imagine a person, who by reading Harry Potter comes to trust the fictional character of Albus Dumbledore. Now, if it turned out that the stories were real and Dumbledore actually existed, the trust awoken by what was (mistakenly) taken to be a fictional story would carry over to the real person of Dumbledore. Likewise, it is plausible that the agnostic can come to trust God in the same way by encountering the stories told about God in the Bible and the Christian community (even though he is unsure whether they are true or fictional). H. H. Price in a similar way connects imagination and religious practice. He claims that agnostic religious practice is "something like what an actor does when he throws himself into his part (Price 1969, pp. 484-85)." He calls it an "imaginative exercise". It is the attempt of the agnostic "to 'put himself in the shoes of' a type of person whose habitual thoughts and feelings are very different form his own (Price 1969, p. 485)." The aim of this exercise is to test the Christian religion, that is, to gain new evidence. Now, there is a difference between emotions based on beliefs and emotions based on the imagination. It is a difference of intensity, but not of kind. The psychologist Paul Bloom states the obvious, that "watching a movie in which someone is eaten by a shark is less intense than watching someone really being eaten by a shark" (Bloom 2011, p. 166). The trust accessible to the agnostic, therefore, seems to be of less emotional intensity compared to the trust accessible to the believer. One might object to this solution in light of the doctrine of the Holy Spirit mentioned above. This solution seems to make possible the existence of Christian faith uncaused by the Holy Spirit, but this solution need not deny the necessity of the Holy Spirit in creating faith. It has often been claimed that it is exactly when the Christian stories are told that the Holy Spirit works. Since the agnostic does not know whether these stories are fictitious or not, it is possible that beliefs are not formed, and he therefore remains an agnostic. Still, it seems possible that an agnostic can come to trust God this way while being agnostic about his existence.
Similarly, we can hope or long for something which we do not believe to be the case. Imagine a father longing for his dead son. He does not believe in his son's existence. Still, he longs for what did exist and maybe also what could have existed. Maybe something similar is possible concerning faith in God. The agnostic can long for God or hope for the existence of God. This longing (or hope) is awakened by the stories of God (the imagination), who God would be if he existed. Longing is a kind of love. We long for something we love. Hope and trust are likewise related. Both hope and trust believe in the good of a person (or object or event). But while trust is more certain and directed at persons physically and socially close to us, hope is less certain and directed objects distanced from us. Therefore, again, it seems possible that an agnostic can have the emotional element of faith (love and trust) despite his agnosticism.

Christian Living Requires Christian Beliefs
The last problem we need to consider is whether agnosticism is compatible with living a Christian life. Stephen T. Davis states, "it would be surprising to find a man basing his life on a proposition about which he is uncertain" (Davis 1975, p. 27). In the same vein, the writer of The Letters to the Hebrews states that "whoever would approach him must believe that he exists and that he rewards those who seek him" (Heb 11:6 NRSV). The basic problem is that without positive belief, the agnostic will lack reasons to base his actions on.
Is it possible to solve this problem? Now, even if a person has no reason at all or all reasons against living Christianly, he could still choose to live a Christian life. This argument shows that compatibility is possible at least in theory. Still, from a psychological perspective, it seems very improbable. We usually do what we believe we have the best reasons to do. If we want to show that agnostic Christian faith is possible, we need to identify some reasons for living a Christian life.
Let us take a look at two kinds of possible reasons. First, he has reason to live a Christian life because living such a life can be a part of a search for more evidence. Praying, going to church, reading the Bible, etc. are ways where one can possibly obtain evidence either for or against belief. The agnostic philosopher, Paul Draper, agrees that agnosticism "calls for religious practice of a certain sort" and that "such practice might ultimately lead to belief by leading first to new evidence" (Draper 2002, p. 206). Of course, this will only make possible temporary compatibility. In time, he will (probably) either believe or disbelieve, or if he finds no evidence able to sway his agnosticism to either side, he will lose his belief that he can gain evidence this way, and therefore lose his reason for continuing to live a Christian life (for an analysis of when it is reasonable to search for evidence, see Swinburne 2005, p. 68).
Second, he has reason to live a Christian life because of the benefits of such a lifestyle. The most famous form of this argument is Pascal's wager (originally presented in Pensé es). The possible good of salvation (and the badness of damnation) is so great that if a person is truly agnostic about the existence of the Christian God, the mere possibility will be a good reason for living Christianly. Whether Pascal's wager is sound has been discussed since his time and the discussion continues today. We will just note that it is a possible reason for agnostics to live a Christian life. Another less disputed good of Christian life is religious meaning. The term "meaning" is sometimes understood to designate the purpose of life. We use the term to designate how we use words (meanings) to interpret life (Baumeister 1991, pp. 15-28). People need meaning to live good lives. The Christian religion gives a set of ideas to interpret and deal with life. One such meaning could be the statement "God loves me." Such a statement could, for example, improve one's sense of self-worth. The problem is that whether this meaning works and enriches one's life depends on whether one believes it is true. The agnostic is unsure, whether it is true, that God loves him. Therefore, the Christian meanings lose their power. But maybe not all its power. As we saw above, we can feel love for characters we know are fictional. Maybe a similar effect is possible here. During prayer or a sermon, we inhabit a world of meaning in the same way as when we read a novel. Even though we are unsure of its truth status, we can feel what it would be like to live in such a world. The third kind of benefit of religious life is purely extrinsic reasons. Such external reasons could be the good of being a part of a (religious) community, the health benefits of religious life (Muller et al. 2001), or, for the pastor, the salary. This last group of reasons seems unaffected by agnosticism (or even disbelief), although the positive health benefits of religious life might change depending on how much of it originates from the positive meaning religion gives to life. These two kinds of possible reasons show that the agnostic can have reasons for living a Christian life. Agnostic Christian faith, therefore, seems possible.
At this point, it might be objected, that these kinds of reasons are theologically problematic. Christian faith is not just doing a set of actions but doing a set of actions for certain kinds of reasons. The actions must ideally be motivated by religious reasons (trust and love of God) and not by external non-religious utilitarian reasons. If this is correct, then Christian acts of faith depend on whether the agnostic can acquire real Christian emotions. That is, if the arguments succeeded in the former section (3B), then it follows that the agnostic can be rightly motivated (by trust and love) in living a Christian life, which some will claim is a requirement for Christian faith.

On the Compatibility of Christian Faith and Evidential Theological Agnosticism
So far, our examination of the compatibility of Christian faith and agnosticism has revealed the possibility of an agnostic having Christian faith although a faith of less emotional intensity. Is the applied definition of agnosticism too strong in comparison with how agnosticism is ordinarily understood? And does this make us blind to certain areas of relevant compatibility?
Agnosticism as we have defined it above is a much narrower category compared to how it is often understood. This is because our definition of evidence as whatever functions as truth-indicating for a person is much broader than is often understood. But, as we saw above, this is because we have defined agnosticism as a doxastic state and not as an epistemological position. So, let us consider for a moment whether Christian faith is compatible with evidential theological agnosticism (EVTA). We will define EVTA as the proposition: There is not enough evidence either for or against God to justify either theism or atheism.
Is it possible to be a doxastic theist while at the same time an evidential agnostic? Can a person believe in God while at the same time believe the proposition EVTA? From a rational perspective, it seems impossible, but people are not entirely rational, and this might open for the possibility of compatibility. The important difference between doxastic and evidential agnosticism lies in the concept of evidence. Above we made a distinction between subjective and objective evidence. So far, we have assumed that evidence is subjective evidence, that is, whatever forms our beliefs. But if we want to consider the compatibility of evidential agnosticism (the proposition EVTA) and Christian faith we need to take evidence to be objective evidence. When we argue about the truth of a proposition, we need to use sound arguments (objective evidence) or else our arguments will not show the truth or falsehood of the proposition. Now, is it possible that a person's doxastic beliefs can be formed by something, which she, on reflection, does not see as constituting objective evidence, but which nevertheless forms her beliefs? The category we have in mind is social evidence, which we take to play a much larger role in our belief-formation than is usually realized. Social evidence is our innate tendency to (consciously or unconsciously) trust the testimony of others-especially the opinion of highly esteemed individuals or the opinion of groups we associate with. Robert Cialdini, a professor in psychology and marketing, states as a psychological fact, that "the greater the number of people who find any idea correct, the more a given individual will perceive the idea to be correct" (Cialdini 2009, p. 119). To some degree, social evidence works at a conscious, critical level. We try to assess the credentials of others as to whether we should believe their testimony or not, but it also works at an unconscious level. To some degree, we unconsciously believe what other people tell us and most of the time we benefit from it (see Code 2020 andHaidt 2008). Now, imagine a philosophically well-educated woman who regularly goes to church. When she reflects on the rational argument for and against EVTA, she finds that the (objective) evidence favors EVTA. Still, she finds herself with (irrational) belief in God because of the social influence of her Christian community. When she reflects on EVTA she becomes an agnostic about God, but when she continues with her daily life, being a part of a Christian community, she slips back into believing in God. On a doxastic level belief and agnosticism are conceptually incompatible. One is replaced by the other. But it might not be unfair to say that she is both an evidential agnostic and a doxastic believer at the same time.
Here it might be objected that compatibility is only possible because her belief is irrational, that is, she believes something which she, on reflection, believes not to be warranted by the (objective) evidence. However, from the perspective of Christian faith, it is not a requirement that belief (or faith) is rational. Of course, rationality is important, maybe it is even ethically required of us, but faith can exist without rationality. And further, since, from a Christian perspective, faith is more important than rationality Christians should seek faith whether or not it is rational. The relative weight between faith and rationality is difficult to assess. If faith is valued unconditionally sectarianism easily follows, which we find problematic. However, faith should not be valued unconditionally, even by Christians. If Christianity is true, then Christian faith is (unconditionally) more important than rationality. But since most of us (if anyone) do not know with certainty whether or not Christianity is true, we cannot assume that faith should always be prioritized over rationality. For these reasons, we argue that Christian faith might be compatible with evidential agnosticism.

On Faith and Fluid Personality
With the last argument, we would like to question a premise underlying how we often talk about faith, namely that faith and belief are quite stable. A person is agnostic and then, over time, he might become a believer, but the reality of our personality is often much more complex. We often float back and forth between belief and disbelief. Sometimes we act based on one set of beliefs, other times on another set of beliefs. The fluidity of our beliefs originates partly from the subtle nature of evidence. Evidence is not as stable a category as is often assumed. Sometimes, we forget the evidence we initially had for a belief, making the belief vulnerable. Sometimes we reinterpret evidence, that is, we make new connections which change the evidential weight of the evidence in regard to certain beliefs. Data can only function as evidence when placed within a larger framework. Sometimes we can see the same belief through different frameworks where it has different amounts of evidence in support. Sometimes the social evidence for a belief changes when we change social environment. This more dynamic or fluid characterization of personality questions the neat label we put on ourselves and others such as "Christian" and "agnostic". A person might switch back and forth between belief, agnosticism, and disbelief, between Christian emotions and other emotions, between Christian living and non-Christian living, depending on the context. This more fluid characterization of human psychology might offer an argument for a certain kind of compatibility between Christian faith and agnosticism. To identify this kind of compatibility we need to distinguish between synchronous and diachronous compatibility. So far, we have assumed that compatibility is synchronous compatibility, that both faith and agnosticism exist in a certain person at the same time. Diachronous compatibility asks whether faith and agnosticism can exist in the same person over time. But how can they be characterized as compatible if they do not exist at the same time? To answer this question, we need to apply Eric Schwitzgebel's helpful distinction between occurrent and dispositional beliefs (Schwitzgebel 2019). An occurrent belief is an actual belief (the kind of belief we have assumed so far), that is, S is in a state of believing p. Dispositional beliefs are those beliefs a person has the disposition to have, that is, the states of believing which will probably occur in S over time given his evidence, background knowledge, intelligence, education, upbringing, identify, personality, social context, etc. Dispositions can also be understood to include all the beliefs which S could have. The faithdisposition which we are interested in is a pattern of belief over time. The importance is not with the now occurrent belief (or non-belief) but with the overall pattern. Even though dispositions are not necessarily occurrent, they are just as real. A man, for example, with a disposition to being angry is still "an angry man" when he is feeling happy (Schwitzgebel 2019). The same goes for belief. A believing woman is still a believer in periods of doubt. The point is, that a person might have the disposition to have Christian beliefs even though his occurrent beliefs are agnostic (or vice versa). In this way, Christian faith and agnosticism can be diachronous compatible even though they might not be synchronously compatible (that is, both occurrent at the same time). This raises the important theological question of whether people characterized by such fluidity also moves in and out of salvation. If this is the case, it would have catastrophic pastoral consequences. The distinction between occurrent and dispositional traits might be a way to solve this problem. God does not judge a person on the basis of her occurrent traits, but on the basis of what dispositional traits she has. Or stated less technical, God's judgement is not decided arbitrarily by whether a person has a good or a bad day, but on the totality of her character.

Conclusions
In this article, we have tried to make a systematic examination of the compatibility of theological agnosticism and Christian faith which is informed by philosophical, theological, and psychological perspectives. A systematic project that takes on such major and complex concepts cannot avoid reducing the complexities of life and reality to something simpler and more manageable but hopefully not in a reductionistic way. It would therefore be fruitful to broaden the study by integrating new and promising fields of research such as belief studies or studies on peoples' understanding of the meaning of life. Agnos-ticism effects lives in a profound way as do all beliefs and thus these fields can be enlightening to our more conceptual study (see e.g., Hicks and Routledge 2015 and Angel et al. 2017).
We began this study by asking whether Christian faith is compatible with theological agnosticism. We then proceeded to define agnosticism as a doxastic state, where a person neither believes nor disbelieves in the existence of the Christian God. Then, we defined Christian faith as having five elements. Christian faith is believing and trusting in the Christian God and acting on this belief and trust. By comparing the two concepts, we have found a complex multilayered relation between them. Agnosticism certainly affects Christian faith at several levels, but we have found reasons to believe that Christian faith can be compatible with both doxastic and evidential agnosticism either in a synchronic or diachronic sense. We, therefore, conclude that an agnostic can have Christian faith, but since the agnostic Christian have a weaker basis for having Christian emotions (like love and trust) and living a Christian life the agnostic position is suboptimal from a Christian perspective. It is implicit in the conclusion, that what is compatible is our definition of Christian faith and agnosticism and not necessarily everything which can be understood by the two concepts.