Why Faith Persists

The Object of Inquiry Is the Empirical Reality of Faith

A philosophical theory's ability to proceed without God as its foundation does not mean that it can disregard human belief in God. The conditions under which a theory holds and the constitution of the empirical world are different questions.

“Releasing Foundational Dependence” has already argued that the core concepts and reasoning of Sustenesis Theory do not require God's existence as a premise. Such theoretical independence, however, does not establish that God does not exist. This paper continues within that boundary. If Sustenesis need not first decide whether God exists, how can it study belief in God and explain why the question continues across generations? (Chen, 2026a)

What must first be acknowledged is the existence of faith, disagreement, and inquiry as empirical phenomena. This does not presuppose the existence of God as their object. What people believe, how those beliefs influence their actions, and how they organise relationships, everyday life, and the transmission of ideas around them are all legitimate objects of study. Even a researcher who regards a belief as absurd cannot thereby deny that people hold it, or substitute a rejection of its content for an explanation of the phenomenon.

“Some people believe that God exists” and “God exists” are different propositions. The truth of the former does not establish the latter; questioning the latter does not make the former false. Similarly, the long history of a question does not show that every formulation of it is sound. It does, however, give reason to investigate how it arises, spreads, and persists.

Faith here means more than assent to a proposition. In the established Sustenesis vocabulary, faith is a sustained commitment through which a subject incorporates an existence, value, or ultimate judgment into the direction of life. Religion is a public and historical structure sustained through doctrine, narrative, ritual, authority, tradition, and community. Personal faith, religious organisations, and questions of faith in public discussion must therefore be distinguished. They are related, but none can stand in for the others. (Chen, Sustenesis Glossary)

This paper principally concerns religious faith directed towards God. Its analysis is not automatically a definition of all religions. Nor does it classify every search for meaning, trust in another person, or commitment to a value as religious faith.

Studying religion as an empirical phenomenon is not an innovation originating with Sustenesis. William James studied religious feelings and personal experience, distinguishing questions of origin from judgments of value. Émile Durkheim examined religion through relations among beliefs, practices, and community. (James, 1902; Durkheim, 1912/1915) What Sustenesis must demonstrate is not that it first discovered faith could be studied, but whether it can organise an explicit structural analysis of faith's formation, maintenance, transformation, and failure.

The starting point is therefore neither that religion must be correct nor that religion is merely an error. It is a reality requiring explanation: how belief in God, together with its acceptance, questioning, and rejection, repeatedly enters empirical life.

The Conditions for Faith Cannot Be Explained by Ignorance Alone

Why human beings develop religious faith is a historical and empirical question larger than the scope of this paper. Sustenesis cannot derive a common origin of all religions from its own concepts. In particular, the initial emergence of a faith, an individual's adoption of it, and a tradition's long-term continuation must not be treated as one question. Conditions that bring a faith into existence need not be those that later sustain it.

Discussing the continuation of religious faith and its questions over millennia does not imply an unchanging entity lasting throughout that period. Continuation occurs through teaching, practice, and reinterpretation. Conceptions of God, organisational forms, and roles in everyday life at different historical moments cannot be assumed identical simply because they bear similar names. (Atran & Henrich, 2010)

Cognitive limitation is a useful starting point for analysis, but not a sufficient explanation. People cannot know all facts, fully predict their futures, or end every inquiry into ultimate grounds merely by understanding some local causes. Yet “people do not know everything” does not entail “people must believe in God”. Not knowing can lead to further investigation, suspended judgment, acceptance of uncertainty, or, under particular conditions, faith.

Religion is likewise not the only response to death, loss, and uncertainty. Intimate relationships, secular ethics, artistic activity, and public life can also help people orient their lives. The conditions in which faith can exist should not be understood as a fixed vacancy that only religion can fill. Rather, in particular circumstances, religious faith can become a practical way of organising experience and action.

One explanation can be proposed for empirical examination. People need not only to understand what has happened, but also to decide how to continue living. Explaining facts and establishing a direction for life are related tasks, but they are not identical. Knowing the specific cause of a loved one's death does not by itself settle how to understand that relationship, bear the loss, or decide what to live for next. Religion may organise these questions through an interpretation of life, responsibility, and ultimate meaning. This does not establish that only a religious interpretation can do so.

The limits of human reason therefore prove neither the correctness of faith nor a lack of rationality among believers. Reason can identify insufficient evidence and reflect on the limits of its own applicability. What requires study is how people understand those limits, and under what conditions they connect uncertain experience to a religious interpretation.

Existing research also suggests that the formation of faith cannot be explained by a single factor. Atran and Henrich's evolutionary analysis connects cognitive tendencies, social learning, ritual practice, and group competition in examining religious commitment. Its relevance here is not that one evolutionary account should be accepted as settled, but that psychological tendencies, cultural forms, and social conditions interact in ways requiring specific investigation. (Atran & Henrich, 2010)

Sustenesis should accordingly treat the needs associated with faith as objects of analysis, rather than presuppose a universal and unchanging instinct. A structure of needs, in this paper, is the particular organisation of emotion, memory, relationships, value judgments, and uncertain experience through which certain subjects sustain a direction for life. Religion can participate in that organisation, but no specific faith follows directly from the existence of a need.

The relation between need and faith may also run in more than one direction. People may enter a faith because of existing difficulties. Alternatively, accepting a doctrine may lead them to understand their lives as requiring salvation, a response to a calling, or the fulfilment of a sacred responsibility. Faith can thus change how a subject formulates a problem, as well as respond to it. Which relation matters more in a particular case cannot be determined in advance.

Faith as a Dynamic Sustenetic Structure

Sustenesis centres on Difference, Constraint, and Sustained Coherence. It asks how a structure forms and maintains sufficient continuity and operational capacity through change. Sustained Coherence means neither immobility nor identical components, and it does not require all participants to agree. It concerns how differences form relationships under constraints that allow them to continue. (Chen, Sustenesis Theory)

In faith, Difference includes distinct subjects, experiences, ideas, roles, and practices. Personal circumstances and religious narratives do not naturally coincide. Doctrine and actual conduct may diverge. Belief and doubt, expectation and reality, communal norms and individual judgment can all become relations that require treatment within a structure. Differences need not all take the form of conflict, but without differentiation there would be no relations among these elements to analyse.

Constraint consists of the specific conditions under which those relations form and continue. Doctrine regulates acceptable interpretations; ritual organises the timing and form of participation; organisations establish roles and authority; family and peer relationships influence transmission. Resources, institutional environments, available knowledge, and personal histories may also constrain how a structure of faith operates. Constraint does not mean suppression alone. It includes rules and conditions that make a practice possible.

Sustained Coherence concerns the maintenance of an identifiable organisation of these differences and constraints through change. A person may reinterpret a doctrine while maintaining a basic commitment to faith. A community may change its membership and activities while retaining recognisable continuity in its practices and relationships. Whether it remains the same structure cannot be settled by its name alone. Actual transmission, organisational relations, and continuity of meaning must also be examined.

Sustained Coherence must consequently not be confused with complete logical consistency. A community's ability to maintain social relationships does not show that all its doctrines are mutually compatible. A person's ability to continue living does not prove every belief they hold. Organisational persistence, personal commitment, propositional consistency, and propositional truth require separate assessment.

Calling faith a sustenetic structure must amount to more than renaming religion. Analysis must identify which relationships constitute the structure, which conditions sustain its operation, and which changes would break its existing continuity. Without such specification, “it exists because it is maintained” is circular.

Nor should a structure be described as an agent with a unified intention. Religious life may be maintained through the different actions of believers, families, organisers, and institutional arrangements. It need not be designed by an overarching subject called “religion”. Structural analysis should reveal how these actions connect rather than hide the processes behind a collective name.

Faith Develops Conditions for Its Own Maintenance Through Life

The relationship between faith and the empirical world is not simply one in which everyday experience externally supports an idea. A more significant possibility is that faith changes actions and relationships after entering everyday life, and those changes become conditions for its subsequent maintenance.

A hypothetical case can illustrate this analytical approach. After experiencing a loss, a person begins participating in a religious community. An interpretation of life's meaning may be the initial attraction. Participation then provides regular meetings, people to talk to, commemorative rituals, and relationships of mutual assistance. These practices change the rhythm of the person's life and connect previously abstract ideas to particular people and events.

The person may subsequently interpret responsibility through this faith, undertake service, and change their conduct. These actions in turn alter how others respond, creating new relationships and a different self-understanding. If the person interprets those changes as signs of the faith's reliability, they may strengthen the initial commitment. Each step requires investigation; the whole process should not simply be reduced to consolation.

In this example, faith organises life and is sustained through the life it helps organise. Previously formed actions and relationships become constraints and conditions for subsequent action. The process is not a one-way movement from idea to conduct, but an ongoing relation among ideas, practices, consequences in life, and reinterpretation.

Social-learning research offers relevant empirical material. In two studies involving participants in the United States, Lanman and Buhrmester found that exposure to models who enacted their religious commitments predicted participants' current theistic belief and religiosity, even after controlling for general religious socialisation. These results support investigating behavioural example. The association alone, however, does not establish causal direction or provide a sufficient explanation of religious persistence across traditions. (Lanman & Buhrmester, 2017)

A more specific research question follows. To what extent does the transmission of a tradition depend on verbal teaching, and to what extent on observing how others practise what they teach? Does commitment change differently when doctrine and conduct support one another than when they clearly diverge? Such questions are closer to testable mechanisms than the general claim that culture perpetuates religion.

This does not justify reducing believers' reasons, understanding, and active judgment entirely to social effects. Scriptural interpretation, philosophical argument, religious experience, and personal reflection may be important grounds of faith for believers. Researchers should examine how those grounds are understood, questioned, and revised, rather than assume all reasons are retrospective rationalisations.

A person's report of religious experience is empirical material that deserves serious consideration. What that experience should be attributed to is a further interpretive question. Acknowledging the experience and its effects does not establish a transcendent source. Conversely, proposing a psychological or social explanation does not automatically show that every associated religious claim is false.

Why the Question of God Continues to Be Asked

The persistence of faith and the persistence of questions about faith are not identical. Someone may cease believing in God while continuing to investigate the question. An organisation may survive while some members doubt its doctrines. An idea may lose its original adherents yet remain an object of philosophy, history, or public discussion.

Persistence therefore concerns several scales. At the level of the subject, inquiry concerns whether faith continues to organise the direction of a person's life. At the communal level, it concerns continuity of practice, resources, authority, and membership relations. At the level of the question, it concerns the conditions under which people continue to formulate, transmit, dispute, and reinterpret it. Persistence at one level does not directly establish persistence at the other two.

“Does God exist?” does not preserve itself as a sentence detached from life. To become a question in practice, it requires people who understand relevant concepts, texts or traditions that make it intelligible, circumstances that renew its personal significance, and occasions for discussion in which different answers meet. The question itself has conditions of formation and maintenance.

Disagreement need not function only as an external obstacle to faith. It may also participate in reproducing the question. Criticism may require believers to explain their reasons again; a new explanation may prompt further challenges. Disputes over education, ethics, or public life can bring a previously private question back into public discussion. This is a proposed mechanism requiring specific evidence, not a claim that controversy invariably sustains faith.

The persistence of contradiction also requires careful interpretation. Different subjects can hold mutually exclusive judgments within the same society. This does not make contradictory propositions simultaneously true. What persists is the existence of distinct judgments and their relationship, not the logical validity of contradiction.

This clarifies the question of why conditions for faith continue. They may involve more than gaps in knowledge that have yet to be filled. Everyday life may continually generate new interpretive needs, while existing traditions influence how those needs are understood. Experience can renew a question, and tradition can supply its form of expression. Their interaction is where Sustenesis should focus its analysis.

These conditions are not a permanent reserve for religion. Specific claims with identifiable empirical consequences remain answerable to evidence and counterevidence. When a claim is refuted, a community may revise its interpretation, reduce its importance, or experience departures, division, or collapse. Explaining why other parts persist does not restore the truth of a refuted proposition.

The fact that a question has long been asked also does not show that it can never be answered. This paper need not declare God's existence undecidable in principle. It requires only that inquiry into faith need not wait for a final verdict on that existential proposition.

Explaining Persistence Also Requires Explaining Change and Failure

If Sustenesis can explain every form of continuation but cannot specify when decline occurs, it has not genuinely entered empirical analysis. The persistence of faith does not guarantee that a particular tradition will survive indefinitely or that faith develops in the same direction in every society.

A Pew Research Center survey of 36 countries, published in 2025, found substantial differences in religious retention and switching. People leave religions in which they were raised, enter a religion from a nonreligious background, or move between religions. Religious identity is the object of the survey; it cannot be equated directly with belief in God. Nor do its samples establish a single global trend. (Lesage, Starr, & Miner, 2025)

Structural analysis must therefore address entry, maintenance, departure, and conversion together. An explanation emphasising communal support must test whether changes in support actually affect faith, rather than look for community only where faith survives. An account emphasising the credibility of examples must examine what happens when teaching and conduct diverge, rather than call every possible outcome the structure's self-adjustment.

Secular relationships and arrangements can assume some roles previously performed by religion without necessarily eliminating faith. But an explanation that identifies these roles as crucial must explain why substitute arrangements weaken some commitments and leave others intact. Different outcomes must correspond to identifiable differences in conditions. The account cannot merely be retold so that the theory is always right.

Internal commitment and external compliance must also be distinguished. An organisation may be stable because members identify with it, but also partly because of exit costs, dependency, or external pressure. Observed participation does not establish doctrinal conviction. Population size, recorded identity, activity participation, personal certainty, and influence on everyday life should be investigated separately where possible.

Longevity does not by itself establish that a structure is beneficial. Relationships that strengthen cooperation within a group do not necessarily benefit those outside it. Atran and Henrich discuss religious commitment in relation to both cooperation within groups and conflict between them. Research should not restrict itself to positive functions of maintenance. (Atran & Henrich, 2010)

Two boundaries must consequently be maintained. Persistence does not prove the reality of the object of faith. Nor does structural persistence establish goodness, legitimacy, or desirability. Explaining how a structure is sustained and arguing that it deserves to be sustained are separate tasks.

The next stage of application research should compare life events, relationship networks, participation, doctrinal understanding, organisational conditions, and changes of faith in specific cases, examining their temporal order where possible. The coexistence of a condition and a belief does not establish causation. Participants' retrospective explanations likewise do not exclude other influences.

This paper proposes conceptual relationships and possible mechanisms for organising inquiry. It is not a completed history of religious origins or a cross-cultural causal model. Sustenesis helps identify what to observe, what to distinguish, and how dispersed factors may connect. It cannot replace the evidence supplied by history, sociology, psychology, and religious studies. If the evidence fails to support a proposed mechanism, that explanation should be revised or abandoned. The assertion that everything is a sustenetic structure must not be used to evade counterexamples.

Conclusion

There is no conflict between Sustenesis Theory's independence from God's existence and its capacity to study belief in God. The former means it need not depend on an existential proposition accepted in advance. The latter means it must still address faith, practices, and questions occurring in the empirical world.

The central claim is that faith is more than a static idea awaiting proof or refutation. As an empirical phenomenon, it can form a continuing organisation through personal commitment, behavioural constraints, cultural transmission, and communal relationships. Its consequences in life can then become conditions of its maintenance, transformation, or breakdown. Questions about faith can likewise acquire continuity through expression, transmission, and dispute.

What Sustenesis must therefore ask is which differences are organised, which constraints allow the relations to operate, what sustains them, and what changes them. Neither the needs associated with faith nor religious traditions should be treated as fixed entities. The conditions for faith should not be presupposed as a vacancy that only the idea of God can fill.

Whether the object of faith is real remains a question to be assessed claim by claim, according to the relevant arguments and evidential conditions. Whatever position a researcher takes, it cannot erase the reality of faith and its questions. Neither can that empirical reality provide a proof that its object exists.

Sustenesis here is a tool for studying how faith forms, persists, changes, and fails. It is not a tool for proving or disproving God. Inquiry need not first answer the ultimate question of faith before explaining why people continually encounter it and how it becomes part of everyday life.

References

• Chen, Geoffrey / 陈朝辉. 2026a. “Releasing Foundational Dependence: The Philosophical Position of God and Soul in Sustenesis Theory” [《解除奠基依赖——维成论中上帝与灵魂的哲学位置》]. Sustenesis Theory official website. Draft 0.1, 29 August 2026.
• James, William. 1902. The Varieties of Religious Experience: A Study in Human Nature. See especially Lectures I and II.
• Durkheim, Émile. 1912/1915. The Elementary Forms of the Religious Life. Translated by Joseph Ward Swain. See especially Book I, Chapter I.
• Atran, Scott, and Joseph Henrich. 2010. “The Evolution of Religion: How Cognitive By-Products, Adaptive Learning Heuristics, Ritual Displays, and Group Competition Generate Deep Commitments to Prosocial Religions.” Biological Theory, 5(1), 1–13.
• Lanman, Jonathan A., and Michael D. Buhrmester. 2017. “Religious Actions Speak Louder than Words: Exposure to Credibility-Enhancing Displays Predicts Theism.” Religion, Brain & Behavior, 7(1), 3–16. DOI 10.1080/2153599X.2015.1117011.
• Lesage, Kirsten, Kelsey Starr, and William Miner. 2025. “Around the World, Many People Are Leaving Their Childhood Religions.” Pew Research Center, March 26. DOI 10.58094/xt5h-b241.