Regress Argument
A regress argument claims that a demand for justification or explanation leads to an unacceptable infinite regress, circularity, or arbitrary stopping point, thereby motivating a particular structure of justification or explanation. In epistemology, it is used to argue that not all justified beliefs can derive their justification from other beliefs.
At a Glance
- Type
- formal argument
- Attributed To
- Aristotle (classical formulation); later systematized in modern epistemology by foundationalists such as Roderick Chisholm
- Period
- Classical Greek philosophy (4th century BCE); modern epistemological usage 20th century CE
- Validity
- valid
1. Introduction
The regress argument is a family of arguments used to show that certain demands for reasons or explanations cannot be satisfied without either going on forever, ending arbitrarily, or looping back on themselves. In its most influential form in epistemology, it concerns the structure of justification: if every justified belief must be supported by another justified belief, then asking “and what justifies that?” appears to generate an endless chain.
This style of reasoning is not tied to one philosophical theory. It functions as a framework within which rival views about knowledge, explanation, and metaphysical dependence position themselves. Foundationalists appeal to it to defend basic beliefs or fundamental entities; coherentists reinterpret it in terms of networks of mutual support; infinitists accept that justification may involve a genuinely infinite series; externalists question the initial demand that justification must always be grounded in accessible reasons. Skeptical traditions use the same pattern to argue that no adequate stopping point is available.
Although contemporary discussions often focus on belief and knowledge, regress arguments also appear in debates about causation, grounding, laws of nature, moral justifications, and the Principle of Sufficient Reason. In these contexts the items in the regress may be events, facts, explanations, reasons for action, or grounding relations rather than beliefs.
Across these areas, two issues recur:
- When, if ever, does an infinite regress become vicious rather than harmless?
- What, if anything, can legitimately terminate a regress of justification or explanation?
The following sections trace the origins of the regress argument, describe its canonical formulation, and survey the main responses and applications across different branches of philosophy.
2. Origin and Attribution
2.1 Classical roots
The standard origin of the regress argument is usually traced to Aristotle, especially in the Posterior Analytics. There he argues that if every piece of scientific knowledge required demonstration from prior known premises, an infinite regress or circularity would follow, making knowledge impossible. He concludes that there must be first principles that are known non-demonstratively.
It is impossible that there should be demonstration of absolutely everything; there would be an infinite regress, so that there would be no demonstration.
— Aristotle, Posterior Analytics I.3
On this basis, many historians attribute to Aristotle the classical formulation of the regress problem, even though he does not use contemporary terminology about “justification” or “basic beliefs”.
2.2 Early skeptical elaborations
Ancient skeptics, especially Sextus Empiricus, develop a more explicitly skeptical version. Among the so‑called Five Modes of Agrippa, one mode is the regress ad infinitum, which is used to show that any proposed justification either continues indefinitely, stops dogmatically, or becomes circular. Sextus does not present this as a route to foundations, but as a way to suspend judgment.
2.3 Modern and early analytic formulations
In early modern philosophy, regress worries appear in discussions of causal chains (e.g., in cosmological arguments) and of knowledge of first principles (for instance, in Descartes’ foundational method). However, the argument was not yet systematized under the label of an “epistemic regress problem”.
The modern epistemological framing—as a general problem about the structure of justification—is often attributed to Roderick Chisholm, especially in Theory of Knowledge (1966). Chisholm formulates the trilemma: given the regress, one must endorse foundationalism, coherentism, or skepticism.
2.4 Attribution in contemporary scholarship
Contemporary authors typically:
| Aspect | Common Attribution |
|---|---|
| Classical insight that not all knowledge can be demonstrative | Aristotle |
| Explicit skeptical regress and trilemma-style pressure | Sextus Empiricus and Agrippan modes |
| Modern epistemic “regress problem” terminology and structure | Chisholm and later foundationalists |
While there is broad agreement on these attributions, some historians emphasize pre-Aristotelian hints (e.g., in Plato) or stress that more recent coherentists and infinitists have substantially reshaped what is meant by a regress of “reasons” or “justifiers”.
3. Historical Context
3.1 Ancient and medieval settings
In antiquity, the regress problem arose against the background of demonstrative science and debates about skepticism. Aristotle’s concern was to secure a model of scientific knowledge that avoided both endless demonstration and vicious circularity. Skeptics employed regress to undercut claims to certain knowledge.
In Hellenistic and later medieval philosophy, similar patterns appeared in discussions of causal dependence and theology. Christian, Islamic, and Jewish philosophers debated whether there could be an infinite regress of causes or whether a first cause or necessary being must terminate explanatory chains. Here the regress argument intersected with cosmological reasoning rather than epistemic justification.
3.2 Early modern developments
During the early modern period, the rise of scientific method and the quest for indubitable foundations (notably in Descartes) gave regress arguments a new role. They underpinned worries about whether sensory beliefs, innate ideas, or clear and distinct perceptions could halt justificatory chains. At the same time, empiricists such as Locke and Hume raised regress-like concerns about justifying inductive and causal beliefs.
3.3 Twentieth-century analytic epistemology
Within analytic philosophy, the regress argument became central through debates about the architecture of knowledge. Logical empiricists, ordinary language philosophers, and later analytic epistemologists grappled with whether observation sentences, experiences, or linguistic practices could serve as foundational stopping points.
From the mid‑20th century onward, the argument was explicitly formulated as the epistemic regress problem. Foundationalists (e.g., Chisholm) used it to argue for basic beliefs. In response, coherentists (e.g., Sellars, BonJour, Lehrer) and later infinitists challenged the assumption that infinite or circular structures are necessarily epistemically defective.
3.4 Late twentieth century and beyond
By the late 20th and early 21st centuries, regress arguments spread beyond epistemology into metaphysics (especially debates about grounding, ontological dependence, and fundamentality) and philosophy of science (e.g., about explanation and theory justification). At the same time, externalist approaches to knowledge questioned the initial assumptions that generated the regress, and pragmatic and contextualist views reconceived when further reasons are legitimately demanded.
Thus, the historical context shifted from purely logical-demonstrative concerns to a broad set of issues about normativity, explanation, and metaphysical structure.
4. The Regress Argument Stated
4.1 Core epistemic formulation
In its canonical epistemic form, the regress argument begins with a requirement:
- A belief is epistemically justified only if it is supported by an adequate reason or evidence.
From this requirement, a sequence is generated:
- Let B₀ be a belief one claims to be justified.
- To be justified, B₀ must be supported by another justified belief B₁.
- For B₁ to be justified, it must be supported by B₂, and so on.
This yields a potential infinite sequence B₀, B₁, B₂, …, where each belief’s justification depends on the next. The argument then notes that there appear to be only a few ways this sequence might be structured:
| Option | Description |
|---|---|
| Infinite Regress | The chain extends backward without end and without repetition. |
| Circularity | At some point a later belief supports an earlier one, forming a loop. |
| Termination | The chain ends in one or more beliefs that are justified without further support. |
Different versions of the argument dispute which of these options are acceptable. Many traditional formulations claim that neither an infinite regress nor circularity can confer justification, leaving only termination in basic beliefs. Others accept one of the alternative options and reject the conclusion that foundations are required.
4.2 General explanatory version
A structurally similar argument appears in non-epistemic contexts. Replace beliefs with facts, events, or explanans:
- For any fact F to be adequately explained, there must be some further fact or principle E₁ that explains F.
- If E₁ itself stands in need of explanation, it must be explained by E₂, and so on.
The same trilemma—infinite regress, circularity, or termination—then arises for explanatory chains. Philosophers disagree about whether explanation, unlike justification, can be adequately grounded in an infinite or circular structure, or whether some fundamental level must be posited.
5. Logical Structure and Form
5.1 Regress as reductio argument
The regress argument is typically cast as a reductio ad absurdum. It starts by assuming some principle about justification or explanation—often that every justified belief requires another justified supporting belief—and then shows that this assumption leads to a problematic regress or circularity. From the undesirability of that outcome, the principle is rejected or qualified.
A schematic form is:
- Assume: For any item X (belief, fact, etc.) to have property P (justified, explained), it must stand in relation R to a distinct item Y that also has P.
- Iterating (1) yields a regress of items X, Y, Z, …
- All possible structures of such a regress are: infinite non-terminating, circular, or terminating in items lacking P or exempted from (1).
- None of these structures is acceptable under the assumption.
- Therefore, the assumption in (1) must be false or restricted.
5.2 Trilemma structure
The argument often takes the form of a trilemma (sometimes associated with Agrippa or Hans Albert’s “Münchhausen Trilemma”):
| Limb | Option | Alleged Problem |
|---|---|---|
| 1 | Infinite regress | Cannot complete or ground justification; normatively defective. |
| 2 | Circularity | Viciously begs the question; offers no independent support. |
| 3 | Dogmatic stop | Arbitrary termination without reasons; violates initial demand. |
Different philosophical positions challenge different limbs of this trilemma: foundationalism accepts a non-arbitrary form of termination; coherentism denies that all circularity is vicious; infinitism denies that all infinite regresses are problematic; some skeptics maintain that no limb is acceptable.
5.3 Variants and parameters
Logically, regress arguments vary by:
- Type of relation R (inferential justification, evidential support, causal dependence, metaphysical grounding).
- Normative standard applied (completeness, independence, accessibility).
- Directionality (backward-looking justification vs. forward-looking explanation).
Despite these variations, their logical form is unified by the idea that repeatedly applying a principle about support leads to an allegedly unacceptable structural pattern, from which a theoretical conclusion is drawn.
6. Types of Regress (Epistemic, Explanatory, Metaphysical)
6.1 Epistemic regress
An epistemic regress concerns chains of justification or reasons for belief. Each belief’s rational status depends on another belief, generating questions about whether such chains can be infinite, cyclical, or must terminate. This is the most widely discussed form in contemporary literature and underpins debates among foundationalists, coherentists, infinitists, and externalists.
6.2 Explanatory regress
An explanatory regress involves chains of explanation:
- Each explanandum calls for an explanans.
- The explanans may itself be in need of explanation.
Examples include regresses of scientific explanation (e.g., explaining laws by higher-level laws) or philosophical explanation (e.g., explaining why there is something rather than nothing). The central issue is whether explanations can be complete or satisfactory if they generate further questions without end, or whether at some point explanation can appropriately bottom out.
6.3 Metaphysical and grounding regress
A metaphysical or grounding regress involves relations of ontological dependence or fundamentality. Here the items are entities, facts, or properties rather than beliefs or explanations:
- Each fact F is grounded in a more fundamental fact G.
- G is in turn grounded in H, etc.
In current metaphysics, such regresses arise when discussing whether there is a fundamental level of reality or whether dependence chains can be infinitely descending. Some argue that certain infinite or circular dependence structures would undermine the very idea of grounding; others maintain that non-well-founded structures could be metaphysically coherent.
6.4 Causal and practical regresses
Some discussions identify additional types:
| Type | Focus | Example Concern |
|---|---|---|
| Causal regress | Chains of causes and effects | Can there be an infinite regress of causes without a first cause? |
| Practical/normative regress | Reasons for action or norms | If each reason requires a higher-order reason, does rational action become impossible? |
Philosophers disagree over whether these are best treated as special cases of explanatory or metaphysical regress, or as autonomous categories with their own standards of adequacy.
7. Vicious vs Benign Regresses
7.1 Distinguishing viciousness
Not all infinite or circular structures are considered problematic. A regress is labeled vicious when it undermines the very property it is supposed to secure (such as justification, explanation, or grounding). A regress is benign when it does not have this undermining effect and may even be expected or explanatory.
Different criteria have been proposed for making this distinction, focusing on:
- Practical impossibility (e.g., finite agents cannot complete an infinite chain).
- Normative failure (e.g., the chain cannot confer the relevant status).
- Metaphysical incoherence (e.g., dependence without a basis).
7.2 Examples of benign regresses
Some philosophers argue that certain infinite chains are unobjectionable:
- In mathematics, an infinite sequence or series may be well-defined and even essential to explanations.
- In metaphysics, an infinite past of events is sometimes taken as coherent.
- In epistemology, infinitists contend that an infinite series of reasons can in principle be available, making the regress benign.
Similarly, not all circularity is deemed vicious. Coherentists maintain that mutual support among beliefs in a network can enhance justification rather than trivialize it.
7.3 Criteria proposed in the literature
Various accounts attempt to mark viciousness:
| Criterion | Idea | Application |
|---|---|---|
| Self-defeating | The regress presupposes what it aims to secure | A chain of justifiers each requiring prior justification never reaches something that can confer justification. |
| Explanatory impotence | Nothing new is explained at each stage | Each step merely restates the explanandum. |
| Dependence without ground | There is dependence but no ultimate basis | Grounding relations form a loop that prevents any fact from being independently grounded. |
Philosophers differ over which criterion is appropriate in which domain. Some maintain a domain-relative approach: what counts as vicious in epistemology may not count as vicious in metaphysics, and vice versa.
8. Foundationalist Responses
8.1 Core foundationalist strategy
Foundationalism responds to the regress problem by positing basic or foundational items—typically beliefs or cognitive states—that are justified or warranted non-inferentially. These items:
- Do not depend on other beliefs for their justification.
- Serve as termini for justificatory chains.
- Support further, non-basic beliefs through inference or other relations.
In this way, foundationalism accepts that infinite regress and circularity would be problematic, but denies that justificatory chains must extend indefinitely.
8.2 Types of foundationalism
Foundationalist views differ over the nature of the foundations:
| Type | Characterization of Basic States | Representative Ideas |
|---|---|---|
| Classical empiricist | Non-inferential sense-data or experiences provide self-evident justification | Early logical empiricism, some sense-data theorists |
| Moderate/weak foundationalist | Basic beliefs are fallible but reasonably prima facie justified by perceptual, memorial, or introspective states | Chisholm, Audi |
| Externalist foundationalist | Foundations are outputs of reliable or properly functioning processes, though the subject may lack access to justifying reasons | Goldman, Plantinga |
Some foundationalists emphasize phenomenal character (how things seem), others conceptual self-evidence, and others yet reliability or proper function.
8.3 Justification of basic beliefs
A key foundationalist issue is how basic beliefs can be justified without initiating a new regress. Proposed answers include:
- Self-justification: some beliefs (e.g., “I am in pain”) are justified by their content and the subject’s state.
- Non-doxastic grounding: experiences or seemings provide justification without themselves being beliefs.
- Externalist warrant: proper functioning in an appropriate environment suffices for justification, even if no reasons are accessible.
Critics contend that such accounts either smuggle in inferential justification (reviving the regress) or make basic beliefs arbitrary. Foundationalists reply by drawing careful distinctions between having justification and being able to justify, or between doxastic and non-doxastic sources of support.
8.4 Role in the broader debate
Foundationalism functions as one limb of the standard trilemma, offering a structured way to halt regress without endorsing skepticism. It provides a contrast class for coherentist and infinitist approaches, and its plausibility often depends on how demanding one takes the initial requirement for reasons to be.
9. Coherentist and Holistic Alternatives
9.1 Coherentism about justification
Coherentism rejects the idea that justification must take the form of a linear chain terminating in foundations. Instead, justification is seen as a property of an entire system of beliefs, where individual beliefs are justified by their mutual support and overall coherence with the rest.
On this view, the regress argument presupposes an overly one-dimensional picture of support. The demand that justification must start from a privileged set of basic beliefs is replaced by the idea that what matters is how well a belief fits into the broader web.
9.2 Holistic support and circularity
Coherentists accept that some circularity is inevitable: beliefs often support each other in complex networks. However, they distinguish between viciously question-begging circles and holistic mutual support. For example, a belief about the external world may be supported by perceptual beliefs, background theories, and inferential principles that in turn are supported by their explanatory success and integration with other beliefs.
Proponents argue that:
- Holistic support can increase the probability or reasonableness of beliefs.
- Demanding linear, non-circular support misconstrues the actual structure of reasoning and theory evaluation.
9.3 Varieties of coherentism
Different forms of coherentism emphasize different coherence relations:
| Version | Emphasis |
|---|---|
| Probabilistic coherentism | Coherence as probabilistic support relations among beliefs |
| Explanatory coherentism | Coherence as explanatory integration and unification |
| System-coherentism | Coherence as consistency, inferential connections, and simplicity of the belief system |
Some coherentists allow for input from experience or observation, but deny that such inputs are justificationally privileged in the way foundationalism suggests. Instead, experiences are treated as causes or constraints on what a coherent system of beliefs can look like.
9.4 Responses to regress worries
Coherentists typically respond to regress arguments by:
- Denying that justification must proceed by tracing a chain backward to a starting point.
- Claiming that epistemic status is determined synchronically by the overall network, not diachronically by historical derivation.
- Arguing that the relevant regress is blocked because the demand for ever-earlier reasons is misplaced within a holistic framework.
Critics question whether coherence without an independent link to truth can yield genuine justification, while coherentists maintain that coherence is intimately connected to truth-conduciveness, explanatory power, or rational acceptability.
10. Infinitism and Acceptance of Regress
10.1 Core infinitist position
Infinitism accepts that justification involves an infinite, non-repeating chain of reasons. Rather than seeking to terminate the regress or transform it into coherence, infinitists argue that the regress need not be vicious if understood properly. On this view:
- No belief is ultimately foundational.
- Justification consists in the availability, in principle, of further reasons without end.
10.2 Motivations for infinitism
Infinitists contend that:
- Foundations appear arbitrary or unjustified if they do not themselves have supporting reasons.
- Coherentist circularity fails to provide the independent support that regress arguments demand.
- An infinite chain best satisfies the intuition that for any reason, one can always ask for a further reason.
They often distinguish between actual possession of an infinite series and potential availability, claiming that finite agents can be justified if they can extend their chain of reasons as needed.
10.3 Structure of infinite chains
Infinitist models impose constraints on the regress:
| Constraint | Purpose |
|---|---|
| Non-repetition | Avoid trivial circularity and ensure genuinely new reasons at each stage. |
| Increasing support | Later reasons must enhance or at least sustain the justificatory status of earlier beliefs. |
| Accessibility in principle | Agents must be able, at least in principle, to supply further reasons upon challenge. |
Under these conditions, the infinite regress is portrayed as a progressive structure of justification rather than a defect.
10.4 Criticisms and replies
Critics argue that infinitism demands an impossible epistemic ideal, since finite agents cannot entertain or articulate infinitely many reasons, and that the view threatens to make justification unattainable. Infinitists respond by invoking notions such as:
- Dispositional possession of reasons.
- Context-sensitive standards: only finitely many reasons are needed in any actual context of challenge.
- Analogies with potential infinity in mathematics.
The debate turns on whether such responses adequately show that accepting a regress yields a benign rather than vicious structure.
11. Externalism and Limiting the Demand for Reasons
11.1 Externalist challenge to the starting point
Externalist theories of justification, such as reliabilism and proper functionalism, challenge the initial assumption that generates the regress: that for a belief to be justified, the subject must have or be able to give additional supporting reasons. Externalists separate:
- The metaphysical grounds of justification (e.g., reliability of the process).
- The subject’s access to those grounds.
On this view, many beliefs are justified even when the believer lacks further beliefs justifying them.
11.2 Blocking the regress
By weakening or denying the requirement that every justified belief be supported by another justified belief, externalists block the regress at its source. A belief may be justified because:
- It results from a reliable cognitive process (e.g., normal vision).
- It is produced by a properly functioning cognitive faculty in an appropriate environment.
In such cases, no further justificatory belief is required. Demands for “yet another reason” can be dismissed as misunderstanding the nature of justification.
11.3 Limiting the scope of “the space of reasons”
Some externalists, and related disjunctivist or virtue epistemology approaches, argue that the space of reasons—where regress arguments operate—is only one dimension of epistemic evaluation. Other dimensions, such as cognitive agency, intellectual virtues, or reliability, need not be structured linearly and so do not give rise to the same regress.
Contextualist and pragmatic approaches similarly limit the regress by holding that:
- The demand for reasons is context-sensitive.
- At some point, further demands become conversationally inappropriate or practically idle.
11.4 Interactions with foundationalism and coherentism
Externalism can be combined with traditional positions:
| Combination | Idea |
|---|---|
| Externalist foundationalism | Basic beliefs are outputs of reliable processes; regress is halted externally, not by self-evident content. |
| Externalist coherentism | Coherence helps track reliability, but not all justificatory status is determined internally. |
In all such combinations, the regress problem is addressed not primarily by the structure of reasons, but by relaxing the internalist requirement that generates the regress.
12. Skeptical Uses of the Regress Argument
12.1 Ancient skeptical deployment
Ancient Pyrrhonian skeptics use the regress of reasons as one of several modes (notably among Agrippa’s Five Modes) to motivate suspension of judgment. When any proposed justification is challenged, it either:
- Leads to a further justification, generating an infinite regress.
- Circles back on itself.
- Ends dogmatically.
Skeptics maintain that no option yields secure knowledge, so rational response is epochē (withholding assent).
12.2 Modern skeptical arguments
In modern epistemology, regress considerations underpin broader skeptical challenges, including:
- Global skepticism about the external world: any attempt to justify perceptual beliefs seems to end in circularity or unjustified assumptions about reliability.
- Skepticism about induction: justifications for inductive reasoning appear to presuppose the very principle they seek to defend, or else regress indefinitely.
- Skepticism about other minds or moral knowledge: demands for higher-order justification apparently cannot be satisfied without unacceptable regress or arbitrariness.
Here, the regress argument functions less as a tool for theory choice (between foundationalism, coherentism, etc.) and more as a means to show that no positive theory can meet all reasonable justificatory demands.
12.3 Skeptical trilemma and its force
Skeptical uses of the regress typically insist that all three options— infinite regress, circularity, and dogmatic stopping—are unacceptable. Unlike foundationalists, skeptics deny that basic beliefs can halt regress without being arbitrary; unlike coherentists or infinitists, they question whether mutual support or infinite chains can genuinely confer justification.
Some skeptics take this to show that knowledge is impossible; others adopt a more modest stance, claiming that certain strong epistemic statuses (certainty, infallibility) cannot be achieved.
12.4 Responses to skeptical regress
Non-skeptical philosophers respond in various ways:
- Foundationalists argue that some stopping points are non-arbitrary.
- Coherentists and infinitists challenge the idea that their respective structures are epistemically impotent.
- Externalists deny that the skeptic’s internalist demand is a requirement of justification.
The ongoing dispute concerns whether regress-based challenges establish genuine epistemic limits, or whether they rest on overly stringent or misapplied standards.
13. Applications Beyond Epistemology
13.1 Metaphysics and grounding
In metaphysics, regress arguments concern chains of dependence among entities or facts. Debates about whether reality has a fundamental level often hinge on whether an infinite regress of grounding relations would be vicious:
- Some metaphysical foundationalists assert that without ultimate grounds, no fact would be adequately explained.
- Others maintain that infinitely descending or even circular dependence structures might be coherent and explanatory.
13.2 Causation and cosmological arguments
In the philosophy of religion and metaphysics of causation, regress arguments appear in discussions of:
- Cosmological arguments: whether there must be a first cause or necessary being, or whether an infinite regress of contingent causes is acceptable.
- Temporal series of events: whether an infinite past is compatible with a complete causal explanation of present states.
Here, the key question is whether explanation or causation demands an originating point, or whether an endless chain could suffice.
13.3 Philosophy of science and explanation
Regress considerations arise in scientific explanation:
- Explaining laws by more fundamental laws may lead to questions about an ultimate explanatory framework.
- Some philosophers argue that explanatory regress is benign when each level provides additional understanding; others seek ultimate theories to avoid an endless “why?” chain.
In methodology, regress worries emerge about justifying scientific methods: attempts to justify inductive practices or methodological norms sometimes appear to presuppose what they defend, suggesting a potential circular or regressive structure.
13.4 Ethics and practical reason
In ethics and metaethics, regress arguments concern reasons for action, moral justification, and normative authority:
| Context | Regress Concern |
|---|---|
| Moral justification | Do moral principles require further moral or non-moral justification ad infinitum? |
| Practical reasoning | Does every reason for acting require a further higher-order reason? |
| Normativity | Can the authority of norms be grounded without circular appeals to normative facts? |
Proposed responses parallel epistemic ones: some posit basic moral principles or ultimate values; others endorse coherentist or constructivist accounts of justification; some accept that normative justification may involve open-ended deliberation.
13.5 Metalogic and rule-following
In the philosophy of logic and language, regress arguments are invoked in:
- Rule-following debates (e.g., Kripke’s reading of Wittgenstein): attempts to justify applications of rules threaten to regress into further interpretations.
- Justification of logical principles: defending inference rules using those very rules suggests circularity, while searching for further justification threatens regress.
These applications illustrate how regress arguments function broadly as a tool to question ultimate justifications across domains.
14. Standard Objections and Replies
14.1 Coherentist objection: circularity need not be vicious
Coherentists object to the premise that all circular justification is illegitimate. They argue that:
- Many epistemic practices involve mutual support.
- A belief can gain justification from its role in a coherent, explanatory network.
Replies from critics maintain that a system must connect to reality in a non-circular way, or else coherence risks isolation from truth.
14.2 Infinitist objection: infinite regress can be benign
Infinitists directly reject the claim that an infinite regress of reasons is vicious. They propose that:
- Justification is a matter of extensibility of reasons.
- Infinite chains are acceptable if each step adds genuine support.
Opponents question whether finite agents can meet infinitist standards and whether potential infinity suffices for actual justification.
14.3 Externalist objection: not all justification requires reasons
Externalists challenge the internalist assumption that fuels the regress. They hold that:
- A belief can be justified by reliability, proper function, or virtue without additional beliefs.
- Demanding further reasons in every case mischaracterizes epistemic evaluation.
Internalist critics worry that this makes justification opaque to the subject and blurs distinctions between lucky true belief and knowledge.
14.4 Skeptical challenge: embrace the regress and suspend judgment
Skeptics accept the regress and maintain that:
- No non-circular, non-arbitrary termination is available.
- Infinite chains are either impossible or normatively inadequate.
They conclude that robust knowledge claims should be abandoned or significantly weakened. Non-skeptical philosophers respond by arguing that at least one of the skeptics’ standards (e.g., requirement of certainty, independence, or completeness) is excessive.
14.5 Debates about viciousness
A further line of objection targets the vicious vs benign distinction itself. Some argue that:
- Criteria for viciousness (e.g., explanatory impotence) are unclear or domain-relative.
- Regress arguments often rely on intuitive judgments about unacceptability that may be theory-laden.
Defenders of regress arguments reply by refining their accounts of viciousness and emphasizing case-based analysis (e.g., distinguishing explanatory from justificatory contexts).
15. Contemporary Debates and Key Figures
15.1 Key epistemological debates
Current discussions focus on:
- The structure of justification: foundationalism vs coherentism vs infinitism vs hybrid theories.
- The internalism–externalism divide: whether regress arguments presuppose that justificatory reasons must be accessible.
- The nature of basic beliefs: whether perceptual, memorial, or introspective states can halt regress non-arbitrarily.
Regress arguments are also invoked in examining higher-order justification, such as whether knowledge requires knowing that one knows, which can generate a higher-order regress.
15.2 Prominent contributors
Some often-cited contemporary figures include:
| Figure | Contribution to Regress Debates |
|---|---|
| Roderick Chisholm | Articulated modern regress problem; defended foundationalism. |
| Laurence BonJour | Developed sophisticated coherentism; critiqued foundationalism and later moved toward modest foundationalism. |
| Keith Lehrer | Advocated coherentism with emphasis on acceptance and systematization. |
| Peter Klein | Principal defender of epistemic infinitism. |
| Alvin Goldman | Leading reliabilist; challenged internalist assumptions behind regress arguments. |
| Alvin Plantinga | Developed proper functionalist externalism; offered externalist foundationalism. |
| Ernest Sosa | Advanced virtue epistemology; engaged with regress in the context of reflective knowledge. |
| Susan Haack | Proposed “foundherentism,” a hybrid of foundationalism and coherentism. |
| Barry Stroud | Analyzed skeptical uses of regress arguments and their implications. |
15.3 Metaphysical and cross-domain debates
In metaphysics, regress arguments are central to discussions of grounding, fundamentality, and ontological dependence, with authors debating whether reality must have a terminating level. In philosophy of science, debates continue about the depth of explanation and whether there can be ultimate explanatory principles.
In ethics and metaethics, contemporary constructivists, realists, and expressivists discuss whether moral justification must terminate in basic values, or whether normative structures can be infinite or holistic.
15.4 Emerging directions
Recent work explores:
- Formal modeling of regress and coherence using probabilistic and network-theoretic tools.
- Contextual and pragmatic treatments of when further reasons are appropriately demanded.
- Comparisons between epistemic, metaphysical, and normative regresses to assess whether a unified theory of regress conditions is possible.
Across these debates, the regress argument remains a common diagnostic tool for testing whether proposed theories can meet their own standards of adequacy without collapsing into infinite chains, circularity, or arbitrary stipulation.
16. Legacy and Historical Significance
16.1 Enduring methodological role
The regress argument has become a standard tool in philosophical methodology. It provides a general strategy for assessing whether a theory’s conditions for justification or explanation can be met non-circularly and non-arbitrarily. This has influenced how philosophers formulate theories, often prompting them to specify:
- What counts as a legitimate stopping point.
- How their view avoids vicious circularity or infinite regress.
16.2 Shaping debates about knowledge
In epistemology, the regress problem has shaped the taxonomy of positions—foundationalism, coherentism, infinitism, externalism, and skepticism—and has structured entire textbooks and research programs. It has also focused attention on the architecture of justification, not merely on its content, affecting how concepts like evidence, reliability, and rationality are analyzed.
16.3 Influence beyond epistemology
Historically, regress reasoning has played a major role in:
| Area | Influence |
|---|---|
| Metaphysics | Framing questions about fundamentality, grounding, and the possibility of infinite chains of dependence. |
| Philosophy of religion | Informing cosmological arguments and critiques thereof. |
| Philosophy of science | Shaping views on the completeness and hierarchy of scientific explanation. |
| Ethics and normativity | Guiding debates on basic moral principles and the authority of norms. |
Its cross-domain applicability has entrenched regress arguments as a general template for probing the adequacy of justificatory and explanatory standards.
16.4 Interaction with skepticism and anti-skepticism
Historically, regress arguments have both fueled skepticism and motivated responses to it. Ancient and modern skeptics have used regress to challenge knowledge claims, while anti-skeptical projects—ranging from Aristotelian foundationalism to contemporary externalism—have often been formulated in explicit reaction to regress-based challenges.
16.5 Ongoing significance
The legacy of the regress argument lies not in a single settled conclusion but in its enduring capacity to organize philosophical problems. It continues to:
- Press theorists to clarify their normative assumptions.
- Reveal tensions between aspirations to completeness and the limits of justification or explanation.
- Provide a shared framework within which diverse traditions—analytic, continental, historical, and non-Western—can articulate and compare their positions on knowledge, reality, and normativity.
As such, the regress argument remains a central fixture in the philosophical landscape, shaping both historical inquiry and contemporary research.
How to Cite This Entry
Use these citation formats to reference this argument entry in your academic work. Click the copy button to copy the citation to your clipboard.
Philopedia. (2025). Regress Argument. Philopedia. https://philopedia.com/arguments/regress-argument/
"Regress Argument." Philopedia, 2025, https://philopedia.com/arguments/regress-argument/.
Philopedia. "Regress Argument." Philopedia. Accessed December 10, 2025. https://philopedia.com/arguments/regress-argument/.
@online{philopedia_regress_argument,
title = {Regress Argument},
author = {Philopedia},
year = {2025},
url = {https://philopedia.com/arguments/regress-argument/},
urldate = {December 10, 2025}
}Study Guide
Regress Argument
A style of argument claiming that a requirement for justification or explanation leads to an infinite regress, circularity, or arbitrary stopping point, thereby motivating a particular structure of justification or explanation.
Epistemic Regress Problem
The challenge of explaining how beliefs can be justified without leading to a vicious infinite regress, circularity, or arbitrary assumptions.
Vicious vs Benign Regress
A distinction between infinite or circular structures that undermine the very property they are supposed to secure (vicious) and those that do not and may be acceptable or even explanatory (benign).
Foundationalism
The view that chains of justification or explanation terminate in basic, non-inferentially justified beliefs or fundamental entities that do not require further support.
Coherentism
The epistemological theory that justification arises from mutual support and coherence within a system of beliefs rather than from linear chains terminating in foundations.
Infinitism
The position that an infinite, non-repeating chain of reasons can, in principle, provide justification, and that no belief is ultimately foundational.
Structure of Justification
The overall pattern or architecture by which beliefs are supported—linear, foundational, coherentist, infinitist, or externalist—at issue in regress debates.
Principle of Sufficient Reason (PSR)
The metaphysical principle that for every fact or truth there is an explanation or reason, often generating regress worries about ultimate grounds.
Why does the demand that every justified belief be supported by another justified belief appear to lead to an infinite regress, and what are the main options for responding to this regress?
On what grounds can we distinguish a vicious regress from a benign one? Apply at least one proposed criterion from the article to an example from epistemology and an example from metaphysics.
Is coherentist circularity genuinely different from the vicious circularity rejected in regress arguments, or is coherentism just a sophisticated form of question-begging?
Does infinitism offer a realistic model of justification for finite agents, or does it set an impossible ideal? How do infinitists try to reconcile infinite chains of reasons with our cognitive limitations?
How do externalist theories such as reliabilism and proper functionalism block or dissolve the epistemic regress problem, and what costs do they incur compared to internalist approaches?
Can the Principle of Sufficient Reason be maintained without generating an unacceptable explanatory or metaphysical regress? If so, how might one limit or reinterpret PSR?
Compare the way regress arguments function in ancient skepticism (e.g., Agrippa’s modes) with their role in contemporary anti-skeptical projects. Do modern foundationalists, coherentists, and externalists genuinely answer the ancient skeptical regress, or do they change the subject?