• home
• about
• new
• essays
• AI
• book

Dr. Joseph Dillard is a psychotherapist with over forty year's clinical experience treating individual, couple, and family issues. Dr. Dillard also has extensive experience with pain management and meditation training. The creator of Integral Deep Listening (IDL), Dr. Dillard is the author of over ten books on IDL, dreaming, nightmares, and meditation. He lives in Berlin, Germany. See: integraldeeplistening.com and his YouTube channel. He can be contacted at: joseph.dillard@gmail.com

Don't Believe ART. Test It

A Response to Frank Visser

Joseph Dillard / ChatGPT

Where Visser's Critique Is Strongest

Frank Visser, in his essay, When Adaptation Becomes a Metaphor, poses what is probably the most important question Adaptive Reorganization Theory presently faces: does “reorganizability” actually explain anything that concepts such as path dependence, institutional inertia, organizational learning, resilience, competence traps, exploration and exploitation, policy lock-in, and feedback failure do not already explain? If it does not, ART is unnecessary. If reorganizability simply means that systems which successfully reorganize are reorganizable, the concept is circular rather than explanatory.

The same problem applies to historical examples. If Rome, Qing China, the Soviet Union, and Kodak are identified as suffering from low reorganizability primarily because we already know what happened to them, they illustrate ART but do not test it. We could tell equally persuasive retrospective stories about many failed systems using other conceptual vocabularies. Likewise, if autonomy, reciprocity, confirmation, and reliability are merely four desirable characteristics of healthy individuals and organizations, ARCR is a collection of reasonable principles rather than an evolutionary hypothesis.

These criticisms identify ART's evidentiary burden. Where I disagree with Visser is in the conclusion he draws from them: that ART is essentially a metaphorical redescription of things we already know. That conclusion goes too far.

ART is not intended as a reductive causal theory that competes with path dependence, organizational learning, realism, liberalism, resilience theory, evolutionary biology, psychology, or institutional economics at their respective levels of analysis. It is better understood as a cross-scale evolutionary framework proposing relationships among adaptation, stabilization, constraint, variation, participation, selection, and reorganization, from which more specific hypotheses can be generated and tested within particular domains.

But that clarification by itself is not enough. Calling ART a “cross-scale framework” merely relocates the problem unless the framework specifies what it adds, how its central concepts are operationalized, how cross-scale similarity would be demonstrated rather than presumed, and what findings would count against ART as a framework rather than merely against one of its local hypotheses. hose are the issues that need to be addressed directly.

Evolution Is Not Deeper Because It Is Older

Visser describes my argument as committing an “evolutionary depth fallacy.” Human beings are biological organisms, he observes, but linguistics is not therefore reducible to cellular biology, economics is not reducible to evolutionary theory, and constitutional law does not become more explanatory simply because natural selection preceded constitutions by billions of years. I agree, but that is not the evolutionary claim ART makes.

ART does not argue that evolution is older than politics and therefore explains politics better. Nor does it argue that political institutions are disguised biological organisms, that civilizations are giant organisms, or that international relations should somehow be reduced to evolutionary biology.

The claim is instead that adaptive systems confront a recurring problem. Successful organization is necessary for persistence, yet successful organization changes the conditions under which subsequent adaptation occurs. Structures that solve yesterday's problems become part of the environment within which tomorrow's alternatives must be generated, evaluated, tested, and selected. ART asks whether this relationship recurs, in appropriately different forms, across biological, psychological, organizational, technological, political, and civilizational scales. That is not a reductionist claim. It is a hypothesis about structural recurrence.

An analogy may make the distinction clearer. Feedback operates in thermostats, organisms, economies, organizations, and interpersonal relationships. That does not mean that marriages are thermostats or that economies can be reduced to engineering. It means that feedback describes a relationship sufficiently abstract to occur in very different systems while taking radically different concrete forms in each. ART makes a similar observation about adaptive reorganization.

Whether that observation is validated or not cannot be decided by pointing out that politics is not biology. Of course politics is not biology, just as an organization is not a brain and a civilization is not an ecosystem. The relevant question is whether these different systems nevertheless confront a formally similar problem involving the preservation of successful organization while retaining capacities required for subsequent reorganization.

If that relationship can be operationalized, if it generates predictions, if those predictions distinguish ART from neighboring explanations, and if they survive serious attempts at disconfirmation, then the evolutionary framing has earned explanatory value. If it cannot, Visser's criticism stands.

ART Does Not Claim to Have Discovered Institutional Rigidity

Visser provides an impressive list of concepts that overlap with reorganizability: path dependence, institutional inertia, organizational learning failure, competence traps, policy lock-in, groupthink, feedback failure, exploration-exploitation imbalance, and loss of adaptive capacity. Quite so. I discuss these relationships extensively in The Adaptive Civilization. ART would be implausible if it bore no relationship to these well-established observations.

The relevant question, however, is not whether components of ART already have names. The question is whether ART identifies a useful relationship among them.

Natural selection, to take an intentionally limited analogy, did not discover variation, inheritance, reproduction, competition, or death. Those phenomena were already observable. The theoretical advance involved specifying relationships among familiar phenomena that generated consequences that were not obvious when each was considered separately.

I am emphatically not comparing the evidentiary status or importance of ART with Darwinian evolutionary theory. ART is nowhere remotely close to possessing that degree of empirical support. The methodological point is simply that showing that the components of a proposed theory are already known does not establish that the proposed relationships among them are redundant.

ART begins with a paradox: adaptation can increase while adaptability decreases. A system can become progressively better adapted to its existing environment while becoming progressively less capable of reorganizing when that environment changes.

Even that observation is not particularly novel. Evolutionary biology, organizational theory, ecology, psychology, and complexity science contain numerous versions of it. ART's stronger claim concerns the mechanism through which this happens and the possibility that this mechanism recurs across scales.

What ART Adds to Existing Frameworks

Here the relationship between ART and neighboring theories needs to be more precise than saying that ART “does not replace them.” ART is best understood as a meta-level process framework with domain-specific hypotheses.

It is not merely a new variable added to realism, liberalism, resilience theory, organizational learning, or psychology. Nor is it simply an umbrella under which all of those theories can be placed. And it is not a competitor that tries to explain away their domain-specific mechanisms. Its proposed contribution is narrower and more specific.

Existing frameworks often explain particular aspects of adaptive success or failure. Realism examines security competition, power, and survival under anarchy. Liberal institutionalism examines cooperation, institutions, interdependence, and credible commitments. Organizational learning examines how institutions acquire, retain, and revise knowledge. Resilience theory asks how systems absorb disturbance and continue functioning. Exploration-exploitation research investigates how systems balance refinement of known strategies with search for alternatives. Path dependence explains how early choices and increasing returns constrain later options.

ART asks a different but overlapping question: How do previously successful adaptations alter the conditions under which future variation, participation, selection, and reorganization become possible?

That makes ART less a replacement theory than a process framework concerned with a particular recursive relationship: the way prior adaptive success changes the selection environment for subsequent adaptation. Its incremental explanatory value would therefore have to be demonstrated in one of three ways.

First, ART should predict outcomes that neighboring theories do not predict. For example, it might identify cases in which conventional indicators of strength, resilience, innovation, or institutional quality remain high while reorganizability is declining.

Second, ART should improve prediction when added to established models. If measures of reorganizability explain additional variance in adaptation to novel mismatch after controlling for resilience, organizational learning, institutional capacity, power, and similar variables, ART has incremental value.

Third, ART should identify interactions among existing variables that neighboring frameworks tend to study separately. For example, it may predict that feedback diversity matters differently depending on reversibility, that autonomy without reliability generates variation without stabilization, or that reliability without confirmation can stabilize maladaptive structures.

If ART does none of these things—if it merely renames path dependence, resilience, learning, and flexibility—then the stronger Visser criticism succeeds.

The More Distinctive ART Hypothesis

The more distinctive proposition can be stated as a recursive sequence:

Successful adaptation tends to become stabilized in structures. Those structures constrain subsequent participation. Changes in participation alter what information enters the system. Changes in available information alter the range of variation that can be generated or considered. Restricted variation changes the population of alternatives upon which selection can operate. Selection therefore increasingly occurs within an environment partly created by previous selections.

The important implication is that a system does not merely become “rigid.” Its previous adaptations can progressively modify the conditions under which future adaptation takes place. This is why ART distinguishes between precipitation and sublimation.

Precipitation is the stabilization of adaptive processes into structures, habits, identities, institutions, incentives, relationships, expectations, and routines. It is indispensable. Without precipitation there could be no organism, personality, organization, scientific paradigm, political institution, or civilization. Everything would have to be rediscovered continuously.

Yet precipitation has a cost. What becomes organized also becomes constraining. Institutions create interests in their continuation. Identities make some alternatives psychologically plausible and others nearly unthinkable. Incentives reward some kinds of information while discouraging others. Expertise increases competence within a paradigm while potentially increasing investment in maintaining the paradigm. Relationships generate commitments, and commitments increase the cost of questioning assumptions on which those relationships depend. A system can therefore continue adapting impressively within its existing organization while becoming progressively less capable of reorganizing that organization.

This distinction helps explain why innovation and reorganizability are not identical. A society may be technologically innovative while politically rigid. A corporation may produce hundreds of new products while remaining incapable of abandoning its underlying business model. An individual may accumulate extraordinary knowledge while remaining incapable of reconsidering a central identity. An artificial intelligence system may generate millions of novel outputs while possessing little ability to reorganize the architecture or selection criteria determining which outputs matter.

ART proposes that successful adaptation can recursively alter the production of variation and the conditions of selection in ways that progressively restrict alternatives to the existing organization. That is a stronger and more falsifiable claim than saying that successful systems sometimes become rigid.

From Precipitation to Sublimation

If precipitation were the entire story, ART would simply be another theory of rigidity. The complementary process is what I call generative sublimation.

Sublimation occurs when previously stabilized constraints are loosened sufficiently for excluded or marginalized information, perspectives, relationships, identities, and possibilities to re-enter adaptive participation. This does not mean that whatever emerges is therefore good, wise, or adaptive. Expansion is not selection. The process requires discrimination.

The general sequence ART proposes is therefore not simply “become more flexible.” It is better represented as movement from relative stabilization through mismatch, loosening of constraints, expanded participation, generation of alternatives, discrimination, behavioral testing, selection, and renewed stabilization.

Psychologically, I describe a closely related sequence as: center ? decenter ? expand ? discriminate ? test ? recenter.

The final step matters as much as the middle ones. Endless decentering is disorganization. Endless expansion is noise. Endless experimentation is instability. Adaptive systems must eventually select and stabilize. The ART problem is therefore not stability versus change. It is how systems repeatedly move between them without allowing either to destroy the conditions required by the other.

Operationalizing Reorganizability

Visser's most serious methodological criticism concerns circularity. If a country successfully reorganizes and we therefore call it reorganizable, while another collapses and we therefore call it insufficiently reorganizable, nothing has been explained. Reorganizability has merely become a label attached retrospectively to outcomes.

Reorganizability therefore has to be measured independently of the outcome it is supposed to predict. This requires more than producing a list of plausible indicators. It requires building and validating an instrument.

A provisional operational model of reorganizability might include at least six separable dimensions.

The first is feedback independence: the extent to which information can reach decision-makers from sources that do not depend upon reinforcing the dominant model.

The second is reversibility: the extent to which experiments and policies can be modified or abandoned without catastrophic cost.

The third is variation capacity: the number and diversity of viable alternatives that can be generated under conditions of mismatch.

The fourth is selection quality: the extent to which alternatives are evaluated against evidence and consequences rather than status, ideology, or institutional self-protection.

The fifth is implementation capacity: the ability to translate selected alternatives into coordinated action.

The sixth is revisability of identity and structure: the degree to which core assumptions, roles, routines, or institutions can themselves become objects of modification rather than being treated as fixed constraints.

These should not initially be collapsed into one score. That would risk building ART's assumptions into the measurement instrument before we know whether the dimensions are empirically distinct.

A serious research program would begin with a larger item pool for each proposed dimension and then examine how those items behave empirically. Factor analysis could test whether the proposed dimensions actually separate. Internal consistency could determine whether items within each dimension cohere. Test-retest reliability could determine whether scores are stable enough to measure something more than momentary mood. Convergent validity could examine relationships with established measures of resilience, psychological flexibility, organizational learning, openness, adaptive capacity, and tolerance for ambiguity. Discriminant validity could test whether reorganizability remains distinguishable from them. The weighting problem should also be solved empirically rather than theoretically.

ART should not assume, for example, that feedback independence and reversibility deserve equal weight. Nor should it assume that the same weighting applies across individuals, corporations, political institutions, and civilizations. Initial studies should preserve the dimensions separately. Predictive models can then determine whether some dimensions matter more under certain conditions, whether interactions matter more than main effects, and whether different scales require different operational forms.

A validated Reorganizability Profile may ultimately prove more useful than a single Reorganizability Index.

That would also fit ART better conceptually. A system might have excellent feedback channels but poor implementation capacity. Another might generate many alternatives but be unable to revise core identity. A third might be highly reversible but unreliable. Treating all three as simply “high” or “low” in reorganizability would conceal precisely the structure ART is trying to investigate.

The operationalization problem therefore remains open, but it need not remain vague. It can be converted into a sequence of measurement problems with clear empirical procedures.

Historical Examples Are Not Tests, but They Are Not Useless

Visser is equally correct that Rome, Qing China, Britain, the Soviet Union, and Kodak do not prove ART. They are historical examples. Historical cases can clarify mechanisms, reveal possible variables, suggest comparisons, and generate hypotheses, but they become dangerous when retrospective explanation is confused with prospective prediction.

Almost every collapse can be redescribed after the fact as a failure of adaptability. A stronger test reverses the procedure. Proposed indicators of reorganizability should be measured independently of later outcomes. Systems confronting comparable classes of mismatch can then be examined to determine whether those indicators predict differences in correction, experimentation, institutional modification, and recovery.

Better still, ART should make prospective predictions. Can it identify organizations or societies that presently appear successful but display measurable declines in reorganizability? Can independent researchers applying specified criteria reach reasonably similar assessments? Do those assessments predict responses to subsequent disruptions? Can ART identify circumstances in which successful adaptation increases rather than decreases reorganizability?

These questions are scientifically more important than attaching ART terminology to the ruins of Rome, but that does not make historical examples useless. Historical cases provide something analogous to a natural laboratory. We cannot rerun the Roman Empire with different institutional arrangements, but we can compare patterns, construct counterfactuals cautiously, examine sequences preceding crises, compare cases that confronted similar pressures but responded differently, and derive variables that can subsequently be tested more rigorously. Historical examples are therefore not evidence sufficient to establish ART, nor are they intended to be. They are one source from which testable ART hypotheses can be generated.

ARCR Is Not Intended as a List of Virtues

Visser also correctly observes that autonomy, reciprocity, confirmation, and reliability all have precedents. Again, they should. The interesting claim is not that nobody previously discovered autonomy or reciprocity, nor is ARCR intended to say that four morally admirable qualities make societies healthy. Its proposed function is more specific.

Autonomy protects relatively independent sources of variation and perspective. Reciprocity allows differentiated agents to interact without either complete isolation or unilateral domination. Confirmation provides mechanisms through which internally generated models encounter correction from evidence, consequences, and perspectives they do not themselves control. Reliability allows sufficiently successful coordination to persist through time so that adaptive achievements do not have to be continuously rediscovered.

Put more abstractly, adaptive systems require some means of generating relatively independent variation, bringing differentiated elements into relationship, discriminating among alternatives, and retaining sufficiently successful arrangements. ARCR proposes that these four functions constitute important elements of the selection infrastructure required for continuing reorganizability. Whether they actually do improve selection is an empirical question.

Four variables may be too few. They may overlap. One may prove unnecessary. Important variables may be missing. ARCR may prove useful at the psychological and organizational scales but not at the biological or geopolitical scales. If so, it should be revised. A theory of adaptation that cannot itself reorganize in response to evidence would provide a rather spectacular demonstration of its own thesis.

Adaptive Space Is Not Necessarily Better Space

Visser makes another important criticism when he observes that expanding adaptive space is not automatically beneficial. ART does not require maximizing possibilities indefinitely.

An organism that generates variation indiscriminately may die. An organization that continually experiments without consolidating successful discoveries cannot function. A government that refuses to make decisions because every possibility must remain available will become paralyzed. A person who continually questions every identity, commitment, value, and relationship may become incapable of sustained action. Expansion is therefore only one phase of adaptive reorganization.

The purpose of expanding adaptive space is not to preserve every possibility. It is to reduce the likelihood that previous adaptations prematurely determine which alternatives are even permitted to appear. Selection remains indispensable.

This is why “adaptive” cannot simply mean open, decentralized, cooperative, flexible, diverse, or pluralistic. Sometimes centralization is adaptive and sometimes decentralization is. Sometimes competition generates essential variation, while sometimes cooperation allows possibilities that competition destroys. Sometimes stability is necessary, while sometimes stability becomes imprisonment.

ART therefore does not propose that one pole of these oppositions is intrinsically superior. It asks whether the system retains mechanisms for recognizing mismatch and, when necessary, generating, discriminating among, testing, selecting, and stabilizing alternatives.

Power and Reorganizability Are Not Opposites

Visser also argues that I contrast power and reorganizability too neatly. Power can enormously increase reorganizability. Wealth can finance experimentation. Military capability can protect a society long enough to reorganize. Industrial capacity can provide redundancy. Technological capability can multiply alternatives. Effective government can implement changes that fragmented political structures cannot.

The ART claim is therefore not that increasing power necessarily decreases reorganizability. Power and reorganizability are partially independent dimensions. A system can possess considerable power and considerable reorganizability, considerable power and little reorganizability, little power and considerable reorganizability, or little of either. The more interesting question is how power is organized and how its accumulation affects the capacity for correction.

Under what conditions does increasing power enlarge adaptive space, finance experimentation, protect diversity, and enable reorganization? Under what conditions does the same accumulation of power insulate decision-makers, suppress contradictory information, protect incumbents, raise the cost of reversal, and make existing structures increasingly difficult to question?

This is not an argument against power. It is an argument for treating the organization of power as one variable affecting reorganizability.

ART Cannot Tell Europe What It Should Do About Russia

While I have my personal opinions, ART cannot deduce from evolutionary principles that Europe should reconcile with Russia. It cannot establish that multipolarity is preferable to unipolarity. It cannot demonstrate that NATO should expand or contract, that sanctions should be imposed or removed, or that any particular peace agreement should be accepted. Those judgments require empirical information and normative criteria that ART alone cannot supply.

What ART can do is ask a different class of questions. Does a geopolitical system preserve sufficiently independent channels of information? Can models of adversaries be corrected by contradictory evidence? Can diplomatic alternatives be investigated without investigation being confused with endorsement? Are agreements sufficiently reliable for experiments in cooperation to become possible? Can policies be reversed when their assumptions fail? Does the system require catastrophe before fundamental assumptions become revisable? Are multiple centers generating alternative strategies, and do mechanisms exist for discriminating among them?

The same questions must be applied to Washington, Moscow, Beijing, Brussels, Kyiv, NATO, BRICS, and other actors if ART is to have analytical credibility. “Adaptive” cannot become a synonym for whichever geopolitical outcome I happen to prefer.

This also clarifies what ART can contribute to international relations. Realism can tell us a great deal about power and security competition. Liberalism can tell us a great deal about institutions, interdependence, and cooperation. ART need not replace either. It asks another question: what happens when the arrangements through which a system has successfully pursued power, security, or cooperation begin restricting its capacity to reorganize when circumstances change?

“It Depends” Is the Beginning of the Question

Visser observes that centralization can help or hurt, decentralization can help or hurt, multipolarity can help or hurt, and power can help or hurt. ART can therefore appear to reduce to the sophisticated observation that “it depends.” But that is where the scientific question begins rather than ends.

It depends on what? ART proposes that an important part of the answer concerns the relationship among existing organization, available variation, participation, information, selection, behavioral testing, and the capacity to reorganize previously successful structures.

A framework that merely says context matters is trivial. A framework that specifies relationships within that context, derives predictions from them, and exposes those predictions to possible disconfirmation becomes scientifically interesting.

ART should therefore not be defended by insisting that it is already an empirically validated theory. It is more useful to distinguish several levels.

At the broadest level, ART is a theoretical framework concerned with recursive constraints on adaptation. Within that framework are theoretical propositions about precipitation, sublimation, narrowing and expansion of adaptive space, innovation versus reorganization, and crisis dependence. Those propositions generate testable hypotheses.The hypotheses generate experiments, comparative studies, prospective predictions, and measurement instruments. Together these constitute a research program. The scientific standing of ART will depend increasingly upon what happens at those lower levels.

The Cross-Scale Claim Requires Stronger Criteria

ART's claim to operate across scales is one of its most ambitious features, and therefore one of the easiest places for analogy to be mistaken for explanation. It is not enough to observe that an individual can become rigid, an organization can become rigid, and a civilization can become rigid, and then conclude that the same mechanism has been discovered at all three levels. Several possibilities have to be distinguished.

The first is mechanistic homology: essentially the same causal structure operates at different scales.

The second is functional analogy: different causal mechanisms produce a similar functional pattern.

The third is statistical regularity: similar relationships appear across scales without implying identical mechanisms.

The fourth is observer-imposed similarity: the analyst has selected descriptions that make different phenomena look alike.

ART should not assume in advance which of these applies. A cross-scale claim therefore needs criteria.

At minimum, the same abstract variables should be definable independently at each scale. The temporal sequence predicted by ART should occur in the same order. The relationship should be observable prospectively rather than only retrospectively. Rival explanations should be considered. Interventions that alter the proposed mechanism should change outcomes in the predicted direction where intervention is possible. And the relationship should survive changes in measurement method.

For example, suppose ART proposes that reduced independence of feedback predicts declining reorganizability. At the individual scale this might be measured through conformity, confirmation bias, or dependence on familiar perspectives. At the organizational scale it might involve hierarchical filtering, career incentives, or suppression of dissent. At the political scale it might involve media concentration, elite insulation, or institutional constraints on correction. The concrete mechanisms differ.

The cross-scale claim would be supported if, despite those differences, reductions in feedback independence consistently predict a narrowing of alternative generation and poorer correction under mismatch. That would not prove identity of mechanism. It would support recurrence of a formally similar relationship.

Cross-scale validity should therefore be established in stages rather than declared globally. A finding at one scale licenses a question at another scale, not an inference. This is an important constraint on ART itself.

The Causal Chain Visser Leaves Behind

Near the end of his critique, Visser strips away what he regards as ART's evolutionary grandeur and leaves what he considers the genuinely interesting part. Successful institutions accumulate commitments. Commitments produce interests. Interests defend institutions. Defended institutions filter information. Filtered information reduces corrective feedback. Reduced feedback makes failure harder to detect, and failure becomes harder to reverse. Eventually the system encounters an environment for which its previously successful adaptations are poorly suited.

I agree that this is one of the central ideas. Where we differ is that I do not understand why this causal chain should be regarded as what remains after the evolutionary framework has been removed. It describes precisely the recursive adaptive process ART is attempting to understand.

Previous adaptations have altered the environment in which subsequent variation and selection occur. The system is no longer selecting from everything that might conceivably work. It is increasingly selecting from what its accumulated structures, incentives, identities, relationships, and models permit it to perceive, generate, legitimate, and test. Previous selections have helped construct the environment within which future selection occurs.

This is what ART calls restrictive precipitation. What initially represented successful adaptation progressively restricts conditions required for subsequent reorganization. The problem is therefore not merely that institutions “become rigid.” It is that successful adaptation can recursively alter both the production of variation and the conditions of selection, making alternatives to the existing organization progressively less available before any explicit decision to reject those alternatives has even occurred.

The complementary process, generative sublimation, involves reopening those constraints sufficiently for previously excluded information, perspectives, relationships, and possibilities to participate again in adaptive selection. This does not guarantee a good outcome. Expansion must still be followed by discrimination, behavioral testing, selection, and renewed stabilization.

Together, restrictive precipitation and generative sublimation describe something more specific than institutional rigidity. They propose a recurring relationship between stabilization and the regeneration of adaptive possibility.

What Would ART Have to Predict That Existing Theories Do Not?

This is the most important challenge in Visser's essay. ART must eventually demonstrate incremental explanatory or predictive value over neighboring concepts.

It should eventually be possible to determine whether measures of reorganizability predict responses to novel mismatch after controlling for wealth, institutional capacity, military power, resilience, political stability, organizational learning, and other established variables. ART should identify circumstances in which conventional indicators of adaptation look favorable while reorganizability is deteriorating. It should distinguish innovation from reorganization empirically. It should predict when highly successful adaptation is likely to narrow rather than expand subsequent adaptive space. It should identify conditions under which systems become increasingly dependent upon crisis for correction. It should predict differences between systems that protect relatively independent feedback and systems in which feedback is increasingly generated by actors whose status depends upon preserving the existing organization, and it should show that adding ART-derived variables improves prediction beyond what existing frameworks already achieve.

If organizational learning measures predict adaptation equally well without reorganizability variables, ART has added no incremental value in that domain. If resilience theory already captures the relevant mechanisms, ART should say so. If an ART variable merely duplicates an established construct under another name, it should either be dropped or explicitly identified as equivalent.

From Civilizations Back to Ourselves

There is, however, another implication of ART's cross-scale claim that is easy to overlook. If ART proposes principles that recur across scales, we do not have to begin testing them with civilizations. We can begin with ourselves. Indeed, we should, but this claim needs an important qualification.

Personal experiments are useful primarily as hypothesis-generating and feasibility-testing exercises, not as strong tests of ART. An individual experimenting on himself or herself is subject to obvious methodological limitations. The experimenter and subject are the same person. Expectations can influence interpretation. There is no blinding. There is usually no control group. Sample size is one. Regression to the mean, placebo effects, spontaneous improvement, demand characteristics, selective memory, and ordinary life events can all mimic intervention effects. None of this makes N=1 research useless. It determines what conclusions can legitimately be drawn from it.

While personal experiments can help clarify variables, identify plausible mechanisms, refine measurement instruments, detect obvious failures, estimate feasibility, generate effect-size hypotheses, and improve experimental design, they cannot by themselves validate ART as a general theory.

If ART has no useful implications at this scale, its claim to cross-scale applicability becomes considerably less interesting. So rather than only debating ART conceptually, I invite Frank and other readers to test some of its propositions.

Experiment One: Test Whether Your Existing Identity Restricts Adaptive Space

Choose a persistent life issue about which you have thought repeatedly but made little progress. Before doing anything else, write down your current explanation of the problem, the available solutions you can imagine, what you believe the other people involved think, and what you expect will happen. This establishes a baseline.

Then deliberately adopt or interview several perspectives that are not identical with your normal waking point of view. In Integral Deep Listening this may involve interviewing a person appearing in a dream, an object or character from a dream, an imagined element associated with the life issue, or another perspective generated through a structured interview. The ART prediction is not that these perspectives will be wiser than you. The prediction is narrower: decentering from habitual identity should increase the number and diversity of alternatives generated.

That can be measured. How many distinct explanations existed before the exercise? How many afterward? How many behavioral alternatives? How many of the later alternatives were genuinely absent from the baseline rather than paraphrases of existing ideas?

The experiment can also fail. If repeated decentering produces no greater diversity of perspectives or alternatives than ordinary reflection, that counts against the claim, but because this is an N=1 experiment, even a positive result would not establish the general claim. It would justify replication.

Experiment Two: Test Whether More Perspectives Produce Better Decisions

Generating alternatives is not enough. Random brainstorming can do that. The stronger ART proposition is that expanding participation can improve subsequent selection when expansion is followed by discrimination and testing.

Take a real decision and record your initial preferred response. Generate alternative perspectives using dream interviews, life-issue interviews, consultation with people who disagree with you, or another structured method. Do not automatically follow any of them. Instead, identify recommendations that recur independently across perspectives. Convert several into small behavioral experiments. Predict the results beforehand. Carry them out and record what happens.

The relevant question is not whether the alternative perspectives “felt profound.” It is whether they generated testable alternatives that produced outcomes your original model failed to anticipate. If they do not, the ART claim receives no support from that experiment.

Experiment Three: Test Psychological Precipitation

Choose a belief or behavioral pattern that has worked well for you over many years. This is important. Do not choose an obvious dysfunction.

Ask what benefits it has produced and what identity, relationships, routines, expectations, or incentives have accumulated around it. Then ask what evidence would now be required for you to conclude that the strategy has become less useful.

Write the criteria down beforehand.

Over the following weeks, actively look for both confirming and disconfirming evidence. This tests a central ART proposition: successful adaptations should be particularly capable of generating structures that protect themselves from revision.

If you find it easier to reconsider strategies that have failed than strategies that have contributed importantly to your success, you have encountered the phenomenon ART calls precipitation at the personal scale. The more interesting question is whether this pattern occurs systematically across people and conditions.

Experiment Four: Test Confirmation

Choose an important belief about yourself, another person, or a recurring conflict. Write down what you believe, your confidence in that belief, and what evidence supports it.

Then ask several people who are sufficiently independent of you to disagree without serious interpersonal cost. Do not ask whether they agree. Ask what they think you are missing, what evidence would change their mind, and what evidence they think should change yours. Afterward, record whether your confidence changed and why.

The ART prediction is not that other people are correct. It is that systems receiving sufficiently independent corrective information should detect some errors that systems relying primarily upon internally generated confirmation do not detect. If independent feedback merely adds noise or systematically worsens judgment, the simple ART expectation requires modification.

Experiment Five: Test Reliability

Many insights fail not because they are wrong but because they never become behavior. Choose one recommendation emerging from the previous experiments and translate it into a small, observable action. Specify when it will occur, what counts as completion, and how often it will be repeated. If useful, involve an accountability partner. Continue long enough to determine whether the intervention alters the original problem.

This tests the transition from expanded possibility to renewed stabilization. ART predicts that insight without reliable behavioral implementation should produce considerably less durable reorganization than insight translated into repeated behavior. Again, that is measurable rather than mystical.

Experiment Six: Compare Insight With Reorganization

Keep a record for a month of insights that feel significant. For each one, record whether it produces a behavioral change, whether the change persists, and whether the original problem subsequently changes. At the end of the month, compare the number of insights with the number of durable behavioral reorganizations.

This directly tests one of the observations from which ART arose: systems may be highly generative while remaining poorly reorganizable. Someone may produce thirty interesting insights and reorganize almost nothing. Another may generate three and reorganize substantially.

If insight generation strongly predicts durable behavioral reorganization without the intermediate processes ART proposes, then ART's distinction between generation and reorganization has been overstated.

Experiment Seven: Test Whether Crisis Is Required for Change

Identify several important changes you have made in your life. For each, ask whether you changed before consequences became severe or only after accumulated mismatch produced a crisis. Then choose one current area in which mismatch is present but crisis has not yet occurred. Establish beforehand what evidence would justify modifying your behavior. Make a small reversible change before circumstances force you to do so.

ART predicts that greater reorganizability should be associated with a reduced requirement for crisis as the trigger for significant correction. Over time, this proposition could be tested much more systematically across individuals and organizations.

Experiment Eight: Test ARCR Rather Than Merely Admiring It

Choose a recurring interpersonal problem and assess it separately in terms of autonomy, reciprocity, confirmation, and reliability.

Are the participants able to express genuinely different perspectives without unacceptable consequences? That addresses autonomy.

Do they influence one another, or does influence largely run in one direction? That addresses reciprocity.

Are assumptions checked against what the other person actually means and against observable consequences? That addresses confirmation.

Are agreements translated into behavior sufficiently consistently to become trustworthy? That addresses reliability.

Then intervene on the dimension that appears weakest and observe whether the interaction changes. The hypothesis is not simply that “more ARCR is good.” More autonomy without reciprocity may produce fragmentation. Reciprocity without autonomy may produce conformity. Confirmation without reliability may generate endless insight without change. Reliability without confirmation may efficiently stabilize a bad arrangement. The stronger ART prediction concerns their interaction.

From N=1 to Pilot Research

Personal experiments should be understood as the first rung of a methodological ladder. The next step would involve repeated N-of-1 designs with standardized measures, preregistered predictions, and independent observers where possible.

After that should come small pilot studies with comparison conditions. For example, participants could be assigned to ordinary reflection, structured decentering, decentering plus ARCR evaluation, or a full ART cycle including behavioral testing and follow-up.

Researchers could then ask not simply whether participants improved, but whether ART-specific components contributed incremental benefit beyond ordinary reflection or established interventions.

Blinding is difficult in psychological interventions, but outcome raters can sometimes be blinded to experimental condition. Expectations can be measured and statistically controlled. Independent observers can rate behavioral change. Predictions can be preregistered. Longer follow-ups can distinguish transient enthusiasm from durable reorganization.

The purpose of personal experimentation is therefore not to bypass scientific research; it is to generate a disciplined path toward it.

What Would Validate the Cross-Scale Claim?

Even if ART performs well in personal experiments, that would not establish that the same processes operate in corporations, governments, civilizations, or biological systems. Evidence has to be built separately at each scale. A stronger cross-scale case would require several things.

Comparable abstract variables would need to be independently operationalized at each scale.

The predicted temporal relationships would need to recur.

Measures taken before major mismatch would need to predict subsequent reorganization.

Rival explanations would need to be tested.

And where interventions are possible, manipulating the proposed mechanism should alter outcomes in the predicted direction.

Only after such evidence accumulates would it become reasonable to speak of a robust cross-scale principle. Until then, cross-scale recurrence should be treated explicitly as a research hypothesis. That is not a weakness to conceal; it is the claim to be tested.

What Would Falsify ART?

This question has to be answered at two levels. Specific ART hypotheses can fail individually. Decentering may fail to increase useful variation. Independent feedback may add little corrective value. ARCR may not predict outcomes. Innovation and reorganization may turn out to be tightly coupled rather than partially independent. Crisis dependence may not correlate with reorganizability. Those failures would require revision of particular propositions.

But Visser's deeper challenge concerns the framework itself. What would count against ART as a whole?

A framework is more difficult to falsify than a single hypothesis, but it is not exempt from empirical risk. ART's core framework would be seriously weakened if several of the following findings converged.

If prior successful adaptation did not systematically alter the conditions of subsequent variation or selection, the central precipitation thesis would fail.

If systems that differ markedly in feedback independence, reversibility, alternative generation, and revisability showed no meaningful differences in their capacity to reorganize under novel mismatch, reorganizability would lose explanatory content.

If established measures of resilience, learning, flexibility, and institutional capacity accounted for outcomes just as well without ART-derived variables, the claim of incremental value would fail.

If ART's proposed sequence—mismatch, loosening, expanded participation, discrimination, testing, reorganization, restabilization—showed no recurring temporal structure across well-designed studies, the process model would be undermined.

If interventions designed to increase proposed reorganizability variables repeatedly failed to improve reorganization, the causal interpretation would weaken.

If cross-scale similarities disappeared once domain-specific mechanisms were controlled, the cross-scale claim would collapse into analogy.

And if researchers could not operationalize ART constructs with acceptable reliability and discriminant validity, the framework would remain too vague to support empirical science.

A sufficiently consistent pattern of such findings would not merely modify ART; it would undermine its central claim that reorganizability represents a distinct and generalizable adaptive problem. That is what framework-level falsifiability has to mean.

ART should not be protected by saying that every failed test simply reflects an incorrectly measured version of “true” reorganizability. If the constructs cannot be measured, distinguished, and used to generate risky predictions, they do not yet function scientifically.

The Burden of Proof

Visser ends by saying that the burden of proof belongs to ART. Agreed. That burden should not be evaded by accumulating additional historical examples, inventing more terminology, or declaring that evolution operates at a deeper level of reality. The next stage requires operationalization, comparison, prospective prediction, attempted falsification, replication, and revision.

Where I differ from Visser is in what follows from that conclusion. A theoretical framework awaiting adequate evidence is not therefore merely a metaphor. It becomes scientifically interesting when it generates propositions that can be operationalized, predictions that can fail, experiments that others can repeat, and findings that would force revision or abandonment.

ART did not arise because I needed or wanted another Theory of Everything. It arose because decades of working with dreams, psychological perspectives, identities, and behavioral change repeatedly confronted me with a puzzle. People can generate remarkable insights without changing. They can understand their problems while continuing to reproduce them. They can become more knowledgeable, sophisticated, accomplished, and successful while becoming increasingly invested in identities and strategies that make fundamental reorganization more difficult.

I eventually began asking whether the same relationship between successful adaptation and declining reorganizability appeared beyond psychology. I found closely related problems described in organizations, institutions, civilizations, artificial intelligence, and evolutionary processes more generally. That observation does not establish ART; it generated ART. And ART became a conceptual context to which my own work was a subordinated possible example at individual scale.

Whether these similarities reflect a genuinely generalizable adaptive process, several superficially similar but causally unrelated processes, or merely my own tendency to perceive the same pattern everywhere is precisely what the research program has to determine.

For that reason, the deepest commitment of Adaptive Reorganization Theory should not be to the survival of Adaptive Reorganization Theory. It should be to whether ART itself can do what it asks of every adaptive system: expose its assumptions to relatively independent perspectives, tolerate disconfirming information, distinguish useful criticism from noise, revise successful but inadequate formulations, test alternatives against consequences, and reorganize when evidence requires it. On that criterion, Visser's critique is not a problem to be defeated. It is valuable corrective feedback.

And that leads to my invitation to Frank and to other readers. Do not simply decide whether ART sounds plausible. Do not accept it because its concepts fit your worldview, and do not reject it because similar ideas already exist elsewhere. Select some of its claims and attempt to make them fail. Test whether decentering actually increases useful variation. Test whether independent feedback improves correction. Test whether successful strategies become harder to reconsider precisely because they have been successful. Test whether insight produces reorganization or merely the experience of insight. Test whether ARCR predicts anything beyond familiar measures of healthy functioning. Test whether small voluntary reorganizations reduce the later need for crisis-driven ones. Then report the failures as carefully as the successes.

Then move beyond N=1. Compare groups. Preregister predictions. Blind outcome assessment where possible. Build instruments. Test incremental validity. Replicate findings. Look for failures.

If ART survives that process, it will become more than an interesting conceptual architecture. If it does not, it should change or disappear.

Otherwise, Visser is correct about a danger that extends well beyond ART. We can spend years debating increasingly elaborate conceptual systems while remaining inside the very problem ART is attempting to describe: generating more ideas without increasing our capacity to reorganize in response to what reality tells us.