But perhaps not entirely for the reason Holmes would have given.
Justice Oliver Wendell Holmes Jr. once offered a warning about the language of rights that remains remarkably relevant. In American Bank & Trust Co. v. Federal Reserve Bank of Atlanta, Holmes cautioned that the word right could become a “deceptive” trap because lawyers could begin with a limited right and quietly transform it into an unlimited one. His conclusion was concise: “Most rights are qualified” (American Bank & Trust Co. v. Federal Reserve Bank of Atlanta 1921, 358).
The deeper problem may lie in where we believe rights come from. If government is the source of a right, then government necessarily has some authority to define that right. What government creates through positive law, government can ordinarily regulate, modify, qualify, or abolish through positive law, subject to whatever higher legal restraints the political system has imposed upon itself.
Natural rights make a fundamentally different claim.
A natural right is not something government gives us. Government may recognize it. Government may protect it. Government may regulate conduct occurring around it. Government may even violate it. But if the right is truly natural, government did not create it and therefore cannot make the underlying right cease to exist simply by changing the law.
That distinction places Holmes in an illuminating conversation with two modern defenders of natural rights: John Finnis in jurisprudence and Randy Barnett in American constitutional theory. Holmes helps us see the vulnerability of legal rights once their content depends upon the state. Finnis and Barnett help explain why rights grounded somewhere beyond the state occupy a different and, I would argue, superior position.
Holmes Was Right About Qualified Rights
The dispute in American Bank & Trust involved the Federal Reserve Bank of Atlanta and smaller Georgia banks. The Federal Reserve Bank claimed a right to present checks for payment. Holmes accepted that such a right existed but rejected the conclusion that possessing that right necessarily authorized every possible manner of exercising it (American Bank & Trust Co. v. Federal Reserve Bank of Atlanta 1921, 358–59).
Holmes used a memorable analogy. A person plainly has a right to give away his own money. Yet giving money to induce someone to commit a crime does not become lawful merely because the giver owns the money (American Bank & Trust Co. v. Federal Reserve Bank of Atlanta 1921, 358).
The point is difficult to dispute.
Saying that I possess a right does not necessarily answer every question about what I may do under the banner of that right.
But Holmes’s jurisprudence pushes the question one step further: Who determines the qualification?
Holmes was deeply skeptical of natural-law reasoning as a source of judicially discoverable legal principles. His jurisprudence helped mark a departure from the older American tradition in which natural law could supply part of the moral architecture of the common law. Andrew Forsyth describes Holmes and the legal realists as breaking from the natural-law tradition and increasingly locating law in actual institutional decisions, social practices, and the exercise of governmental power rather than in a coherent body of antecedent moral principles (Forsyth 2019, 125–45).
That shift matters enormously.
If the state is ultimately responsible for defining what counts as a legal right, the statement that “most rights are qualified” acquires an additional meaning. Rights are qualified because the legal authority that recognizes them also defines their boundaries.
And whatever legal authority can define, it can potentially redefine.
The Problem with a Right Government Gives You
Suppose a legislature creates a statutory entitlement to a particular government benefit.
That entitlement is certainly a legal right while the statute remains in force. Courts may enforce it. Officials may be legally required to honor it. A citizen may properly say, “I have a right to this benefit.”
But the legislature that created the program may generally amend the statute. A later legislature may alter eligibility requirements. It may reduce the benefit. It may replace the program. It may abolish it altogether, provided that doing so complies with other controlling law.
The reason is straightforward.
The right depends for its existence upon positive law.
Constitutional rights are considerably more difficult to alter because ordinary legislation is subordinate to constitutional law. Yet even here the distinction between the legal recognition of a right and the source of the right itself becomes important. A constitutional provision is positive law and may ordinarily be altered only through the constitutional process established for amendment. The natural-rights tradition makes a more radical claim: some rights would continue to exist as moral rights even if a constitution failed to recognize them.
This gives us two very different propositions:
Government has declared that you possess this right.
And:
You possess this right, therefore government is obligated to respect it.
Those sentences may produce the same result when government behaves justly. Jurisprudentially, however, they run in opposite directions.
The first makes government the source.
The second makes government the subject of an obligation.
That difference is the heart of natural-rights theory.
Finnis: Law Does Not Create the Moral Order It Governs
John Finnis provides one of the most important modern accounts of natural law and natural rights. His argument is not that human beings possess an unlimited collection of individual preferences that government must recognize as rights. Finnis instead grounds rights within practical reason, objective human goods, justice, human flourishing, and the common good (Finnis 2011).
For Finnis, positive law performs indispensable work. Human communities need legal institutions to coordinate conduct, establish determinate rules, resolve disputes, and pursue common goods that cannot be secured by isolated individuals acting independently (Finnis 2011). His natural-law theory therefore does not treat government as unnecessary or legal rules as presumptively illegitimate simply because they restrict individual choice.
But positive law does not create the entire moral universe within which it operates.
Finnis argues that natural-law reasoning can ground genuine human rights, including rights whose normative force does not depend simply upon governmental recognition (Finnis 2015, 199–225).
This produces an important distinction for our discussion of Holmes.
Natural rights may still have boundaries.
They are not necessarily absolute permissions to do anything associated with the subject of the right. My right to property does not entail a right to use my property to murder another person. My liberty does not entail a moral entitlement to enslave someone else. The moral order that grounds rights also structures their proper exercise.
So Finnis need not disagree with Holmes that rights are often qualified.
The disagreement concerns something deeper.
Who, or what, supplies the qualification?
For Holmes, legal institutions ultimately determine the operative legal boundary. For Finnis, positive law operates within a moral order that government did not create. Government can specify many of the practical boundaries of legal rights, but government itself remains subject to standards of justice external to its own will (Finnis 2011; Finnis 2015, 199–225).
That changes the meaning of a qualification.
A natural right may be limited because the right itself exists within an objective structure of justice.
That is very different from saying a right is limited merely because government has decided how much of the right its citizens will be permitted to exercise.
Barnett: Rights “Retained,” Not Rights Granted
Randy Barnett brings this argument directly into American constitutional law.
The Ninth Amendment contains one of the most intriguing phrases in the Constitution: it speaks of rights “retained by the people.”
Barnett takes that language seriously.
His account of the Ninth Amendment argues that the Constitution assumes the existence of rights that were not created by their enumeration in the constitutional text. On Barnett’s reading, the people possessed natural rights before the Bill of Rights was adopted, and enumerating some rights was not supposed to diminish the status of those that remained unenumerated (Barnett 2006, 1–82).
The word retained matters.
You cannot retain something that you never possessed.
The constitutional language therefore fits naturally with a theory in which government does not grant all rights to the people. Instead, people enter political society already possessing rights, while constitutional government receives powers.
That reverses the usual modern assumption.
Under the government-as-source model:
Government possesses authority and grants citizens certain protected rights.
Under the natural-rights model:
People possess rights and grant government certain limited powers.
Barnett builds much of his constitutional theory around that distinction. In Restoring the Lost Constitution, he argues for a “presumption of liberty” in which government bears a substantial burden when it restricts the rightful exercise of individual liberty. His theory connects constitutional legitimacy to the protection of rights retained by individuals rather than treating individual freedom as merely the remainder left after government has exercised its powers (Barnett 2013).
This provides the American constitutional counterpart to Finnis’s jurisprudential argument.
Finnis asks why rights bind government at all.
Barnett asks what the Constitution should look like if we take seriously the proposition that people possess rights before government acts.
Natural Rights Are Different Because Government Cannot Repeal Their Source
This leads to what I think is the strongest argument for privileging natural rights.
Consider two rights.
The first exists because the legislature enacted Statute A.
The second exists because every human person possesses a moral claim that precedes the state.
The legislature repeals Statute A.
The first right may cease to exist.
Now suppose the legislature announces that the second right no longer exists.
Something very different has happened.
If natural-rights theory is correct, the legislature has not abolished the right. It has violated it.
That distinction is everything.
A government can repeal a statutory entitlement because the statute constituted the entitlement.
A government cannot repeal a natural right in the same sense because governmental action did not constitute the right in the first place.
Government can refuse to recognize it.
Government can violate it.
Government can punish those who attempt to exercise it.
Government can persuade courts to ignore it.
Government might even construct an entire legal regime contrary to it.
But none of those actions answers the moral question of whether the person still possesses the right.
Finnis’s natural-law jurisprudence provides precisely this possibility of evaluating positive law from a moral standpoint not reducible to the positive law itself (Finnis 2011; Finnis 2015, 199–225).
That is the great advantage of natural rights.
They give us somewhere to stand when government itself is wrong.
The Common Law Once Understood This Relationship Better
There is an important historical dimension to this argument.
American law did not always draw a sharp boundary between positive law and natural law. Andrew Forsyth’s history of American common law shows that jurists such as Joseph Story could understand natural law as supplying principles, duties, rights, and limits while common-law doctrine gave those principles increasingly specific legal form (Forsyth 2019, 70–104).
Over time, natural law became less explicit in American legal reasoning. It was increasingly absorbed into common-law doctrine and later challenged by intellectual movements that emphasized law as a product of human institutions rather than an expression of an objective moral order (Forsyth 2019, 70–145).
Holmes stands near the far end of that transition.
This is why placing Holmes beside Finnis is so revealing.
Holmes says: Most rights are qualified.
Finnis can answer: Of course. But government is not necessarily the source of either the right or its ultimate moral limits.
Barnett can then add an American constitutional response: The Constitution itself speaks of rights the people “retain,” suggesting that at least some rights are conceptually prior to governmental enumeration (Barnett 2006, 1–82).
The debate therefore is not really between people who believe rights have limits and people who believe rights are unlimited.
That would be too simple.
The important debate concerns the source of the right and the source of its limits.
Natural Rights Can Be Qualified Without Becoming Government Rights
This qualification is essential to the argument.
If I simply say, “Natural rights are absolute, while government rights are qualified,” I create a claim that is difficult to defend.
Natural rights themselves may be structured by duties owed to other persons, competing rights, justice, and the common good. Finnis certainly does not understand natural rights as limitless individual permissions detached from moral obligation (Finnis 2011).
The stronger proposition is this:
Natural rights are superior to government-created rights not because they are necessarily unlimited, but because their normative existence does not depend upon the government that is being asked to respect them.
That distinction allows us to agree with Holmes without surrendering the natural-rights tradition.
Yes, most rights may be qualified.
But qualification raises a second question that Holmes’s sentence does not answer:
Qualified by what?
If the answer is merely “whatever government decides,” then the right ultimately rests on governmental permission.
If the answer is that a right is bounded by the equal dignity and rights of others, by principles of justice, by the nature of the good being protected, or by requirements of the common good, then the qualification may arise from the same moral order that generated the right in the first place.
Those are profoundly different conceptions of a right.
The Difference Between a Right and a Permission
This brings us to the practical danger of locating the source of rights exclusively in government.
If government gives me liberty today and may withdraw it tomorrow whenever the proper political procedures have been followed, it becomes difficult to explain why I possessed a right rather than a particularly secure governmental permission.
Natural-rights theory provides an answer.
A right is capable of standing against the state precisely because its moral authority does not depend entirely upon the state.
Barnett expresses this insight constitutionally through rights “retained by the people” and his presumption of liberty (Barnett 2006, 1–82; Barnett 2013).
Finnis expresses it jurisprudentially by locating human rights within an account of practical reason and human good that allows positive law itself to be evaluated according to standards of justice (Finnis 2011; Finnis 2015, 199–225).
Holmes reminds us that rights require boundaries.
Finnis and Barnett remind us that government need not be the author of everything it is obligated to respect.
And that may be the most important distinction of all.
Holmes Was Right, but Natural Rights Matter Because He Was Right
Holmes’s statement that “most rights are qualified” can initially sound like an argument against strong conceptions of individual rights.
I think it can point in the opposite direction.
Precisely because positive legal rights can be qualified, altered, and sometimes withdrawn by the legal institutions that define them, we have reason to ask whether there are rights whose authority does not originate in those institutions.
Natural-rights theory answers yes.
That does not mean natural rights have no limits. It means their existence does not depend upon legislative generosity, administrative discretion, or even constitutional enumeration.
The distinction can be stated simply:
A government-created right exists because the law gives it to us. A natural right makes a claim upon the law because we possess it already.
That is why the source of rights matters.
If government is the source of a right, then government possesses an extraordinary degree of authority over its definition.
If the right precedes government, the relationship reverses.
The question is no longer simply, What rights has government decided to give us?
The question becomes, What rights must a just government recognize because they were never government’s to give away in the first place?
Holmes helps us see why rights are often qualified.
Finnis and Barnett help us see why some rights must stand on firmer ground.
Brandon L. Blankenship is an assistant professor, continuing legal education presenter, and business educator. He and his wife Donnalee live on their hobby farm south of Birmingham, Alabama.
We often talk about rights as though the word settles the argument.
“I have a right to do this.”
“You have no right to stop me.”
“The government cannot interfere with my rights.”
But the word right can hide as much as it reveals. More than a century ago, Justice Oliver Wendell Holmes Jr. warned about exactly this problem. In American Bank & Trust Co. v. Federal Reserve Bank of Atlanta, Holmes wrote that the word “right” is “one of the most deceptive of pitfalls” because it is easy to begin with a qualified right and end by treating it as an absolute one. His conclusion was simple: “Most rights are qualified” (American Bank & Trust Co. v. Federal Reserve Bank of Atlanta 1921, 358).
Holmes was not saying that rights are meaningless. He was saying that identifying a right is often the beginning of legal analysis, not the end.
What Does It Mean to Say That a Right Is “Qualified”?
Suppose someone says, “I have a right to use my property.”
That may be true. But it does not necessarily follow that the person may use the property in any way, at any time, for any purpose, regardless of the harm caused to others.
The same is true of many familiar rights. A person may have a right to enter a contract, own property, speak freely, conduct business, or transfer money. Yet the law still asks questions about how that right is exercised and whether its exercise violates the rights of someone else.
That was Holmes’s point.
The case before the Supreme Court involved banks and the Federal Reserve Bank of Atlanta. The Federal Reserve Bank argued, in effect, that because it had a legal right to present checks for payment, it could present large numbers of checks at once regardless of the purpose or effect of doing so (American Bank & Trust Co. v. Federal Reserve Bank of Atlanta 1921, 358).
Holmes rejected that logic.
A right to present a check for payment did not automatically become an unlimited right to use the presentation of checks as a means of coercing or injuring another business. The problem was not the existence of the right. The problem was treating a limited legal right as though it carried no boundaries (American Bank & Trust Co. v. Federal Reserve Bank of Atlanta 1921, 358–59).
Holmes illustrated the point with a simple example. A person may ordinarily have the right to give money to another person. But if the payment is a bribe, the fact that the money belongs to the giver does not make the transaction lawful (American Bank & Trust Co. v. Federal Reserve Bank of Atlanta 1921, 358).
The right is real. The exercise of the right is still subject to legal limits.
The Word “Right” Can Hide Important Questions
Modern legal philosophy helps explain why Holmes’s warning remains useful.
When lawyers use the word right, they may be describing several different things. A right might mean that a person is free to act. It might mean that another person has a duty toward them. It might mean that a person has the legal power to change a relationship. It might also mean that the law protects a person against someone else changing that relationship (Wenar 2005, 223–52).
That means the statement “I have a right” may be incomplete.
A better set of questions is:
What kind of right?
A right against whom?
A right to do what?
Under what circumstances?
Subject to what limits?
These questions matter because rights exist within relationships. My rights interact with your rights. Property rights interact with nuisance law. Contract rights interact with fraud rules. Religious liberty interacts with other legal obligations. Freedom of speech interacts with doctrines governing threats, defamation, and other legally significant forms of expression.
The existence of a right therefore does not necessarily answer the question of its scope.
Holmes and the Common Law
Holmes’s observation fits comfortably within one of the central habits of common-law reasoning.
The common law usually develops by drawing boundaries between competing claims. Judges ask where one person’s legal freedom ends and another person’s legally protected interest begins.
That means common-law rights are rarely isolated propositions. They exist within a network of duties, defenses, remedies, exceptions, and competing rights.
An owner has property rights, but neighbors also have rights.
A business may compete with another business, but not every form of competition is necessarily lawful.
A person may make an agreement, but the law may refuse to enforce agreements involving fraud, duress, illegality, or other recognized defects.
Historically, American common law also developed alongside natural-law reasoning. Lawyers and judges did not necessarily treat natural law as a simple list of absolute rules. Natural-law principles could provide broader ideas about justice, duty, reason, and human flourishing, while common-law doctrine supplied more specific rules for applying those principles to actual disputes (Forsyth 2019, 70–104).
This is important because it shows that the idea of qualified rights did not originate with Holmes.
Natural Rights Do Not Have to Mean Absolute Rights
At first glance, Holmes may appear to be in tension with the natural-rights tradition.
Natural-rights language suggests that some rights exist because of what human beings are, not merely because a government has chosen to recognize them. If that is true, one might assume that natural rights must also be absolute.
But that conclusion does not necessarily follow.
Natural-law theorists have long understood rights within a larger framework of justice, practical reason, human dignity, and the common good. A natural right may be fundamental without implying that every imaginable use of that right must be legally protected (Finnis 2011).
Consider property again.
A natural-rights theory might strongly protect the human capacity to own property. But it does not necessarily follow that ownership gives a person moral permission to use property to poison a neighbor’s land, defraud another person, or deliberately cause unjust harm.
The right itself may be fundamental. Its exercise still occurs within a moral and legal order.
This is one place where natural-law thinking and common-law reasoning can reinforce each other. Natural rights can help explain why certain interests deserve protection. Common law can help define how those interests operate when they encounter the legitimate interests of other people.
Where Holmes Breaks from the Natural-Rights Tradition
Holmes nevertheless differs significantly from traditional natural-law thinkers.
His disagreement concerned more than whether rights have limits. Holmes was skeptical of the idea that judges could begin with abstract concepts such as liberty, property, or natural right and logically deduce the answer to difficult legal disputes.
He wanted courts to examine what legal rules actually did in society and how competing interests interacted in practice. His work helped influence the broader movement away from nineteenth-century legal formalism and toward the more pragmatic approaches that later became associated with legal realism (Cohen-Eliya and Porat 2010, 263–86).
This marked an important transition in American jurisprudence.
Natural-law reasoning remained influential throughout much of American legal history, but lawyers increasingly began to understand law as the product of human institutions, judicial decisions, statutes, and historically developed legal practices rather than as a body of principles waiting to be discovered through reason alone (Banner 2021).
Holmes stands near the center of that transition.
So although Holmes and a natural-law theorist might agree that rights have limits, they may disagree about where those limits ultimately come from.
A natural-law thinker might say that rights are bounded by justice, human goods, duties to others, and the common good.
Holmes was more likely to look at the legal system itself and ask where courts had drawn, or should draw, the boundary between competing interests (Forsyth 2019, 125–45).
The Later History of the Case Matters
There is another important lesson in the history of American Bank & Trust.
When the dispute returned to the Supreme Court after further factual development, the courts concluded that the Federal Reserve Bank had not actually acted with the unlawful coercive purpose originally alleged. The Federal Reserve Bank was therefore permitted to continue presenting checks even though its conduct created economic pressure on the smaller banks (American Bank & Trust Co. v. Federal Reserve Bank of Atlanta 1923, 648–49).
That result helps clarify Holmes’s point.
He was not saying that whenever exercising a right harms someone else, the right disappears.
He was saying that the bare statement “I have a right” does not settle the question.
Courts still have to determine the scope of that right, the circumstances in which it is being exercised, and whether the conduct remains within lawful boundaries.
Sometimes the answer will be yes.
Sometimes it will be no.
But the answer cannot be reached simply by repeating the word right.
Why This Still Matters
Holmes’s warning remains valuable because modern public debate often treats rights as rhetorical trump cards.
Once someone invokes a right, further discussion can sound like an attack on the right itself.
But that is often a mistake.
A mature legal system can strongly protect rights while still asking difficult questions about their boundaries. In fact, defining those boundaries is one of the principal things law does.
Natural-rights theory helps us ask why certain human interests deserve protection in the first place. The common law helps translate those interests into legal relationships, rules, remedies, and limitations. American constitutional law adds another layer by placing some rights beyond ordinary political interference while still requiring courts to determine their scope.
Holmes’s warning therefore deserves to be remembered.
The danger is not believing in rights too strongly.
The danger is using the word right too loosely.
A legal right has a holder. It has an object. It operates against particular persons or institutions. It exists within particular circumstances. And it has boundaries.
Understanding those qualifications does not necessarily weaken a right.
Often, it is what makes the right legally meaningful.
Brandon L. Blankenship is an assistant professor, continuing legal education presenter, and business educator. He and his wife Donnalee live on their hobby farm south of Birmingham, Alabama.
American Bank & Trust Co. v. Federal Reserve Bank of Atlanta. 1921. 256 U.S. 350.
American Bank & Trust Co. v. Federal Reserve Bank of Atlanta. 1923. 262 U.S. 643.
Banner, Stuart. 2021. The Decline of Natural Law: How American Lawyers Once Used Natural Law and Why They Stopped. New York: Oxford University Press.
Cohen-Eliya, Moshe, and Iddo Porat. 2010. “American Balancing and German Proportionality: The Historical Origins.” International Journal of Constitutional Law 8 (2): 263–286.
Finnis, John. 2011. Natural Law and Natural Rights. 2nd ed. Oxford: Oxford University Press.
Forsyth, Andrew. 2019. Common Law and Natural Law in America: From the Puritans to the Legal Realists. Cambridge: Cambridge University Press.
Wenar, Leif. 2005. “The Nature of Rights.” Philosophy & Public Affairs 33 (3): 223–252.
Translating complexity into progress requires more than simplifying a difficult problem or applying specialized expertise to one part of it. Complex human problems combine uncertainty about causes, disagreement about values, fragmented institutional authority, and practical difficulty in changing established behavior. They therefore resist linear models in which researchers identify a solution, legal experts approve it, ethicists evaluate it, and practitioners subsequently implement it. Progress depends instead on a recursive process in which research, legal reasoning, ethical judgment, and practical action continually inform and correct one another (Alford and Head 2017; Van de Ven 2007).
Within an approach centered on aligning people, purpose, and practice, research clarifies what is happening and why, legal analysis determines what institutions may and must do, ethical analysis evaluates what they ought to do, and implementation reveals what can work under actual conditions. None of these forms of inquiry is sufficient by itself. Research without implementation may produce knowledge that remains detached from institutional capacity. Legal analysis without empirical grounding may identify formal authority while overlooking how rules operate in practice. Ethical analysis without institutional specificity may articulate compelling principles without determining how competing obligations should be reconciled. Implementation without research, law, or ethics may generate activity while reproducing the conditions that created the problem.
The central task is therefore not to reduce complexity until it becomes simple. It is to organize complexity so that participants can make defensible decisions, undertake bounded action, learn from consequences, and progressively improve the system. This approach replaces the expectation of a final solution with a disciplined account of progress. Complex problems differ in the sources and degrees of their difficulty, so interventions should be matched to the particular combination of knowledge uncertainty, institutional fragmentation, and stakeholder disagreement present in each case (Alford and Head 2017).
Beginning with the Problem Rather than a Preferred Solution
The translation process begins with rigorous problem formulation. Organizations frequently define problems in terms of the program they hope to create, the funding they hope to obtain, or the activity they are prepared to perform. This reverses the proper sequence of inquiry because it allows an available solution to determine which features of reality receive attention. Engaged scholarship treats problem formulation as a foundational activity that should be conducted with people who study the problem, administer the relevant systems, implement services, and experience the consequences (Van de Ven 2007).
A well-formulated problem statement identifies the human condition requiring attention, the population affected, the institutional processes shaping that condition, and the practical decisions that the inquiry must inform. It also distinguishes the presenting problem from the mechanisms that sustain it. A low program-participation rate, for example, may reflect inadequate communication, burdensome eligibility rules, distrust of the administering institution, transportation barriers, inconsistent frontline practices, or a program that does not address what participants value. These possibilities imply different research questions, legal concerns, ethical judgments, and implementation strategies.
Research should therefore construct an explanatory account rather than merely collect information. Quantitative evidence can establish the scale, distribution, and correlates of a condition, while qualitative evidence can clarify processes, meanings, institutional routines, and experiences that aggregate measures obscure. Administrative records can reveal patterns in system performance, but those records also reflect prior decisions about what institutions count, classify, and preserve. Community knowledge and practitioner experience may identify failures that formal data systems have never been designed to register. Engaged inquiry becomes more credible when these forms of knowledge are placed in conversation rather than arranged in a hierarchy that automatically privileges one source (Van de Ven 2007).
This process should result in a working causal model. The model identifies how policies, organizational routines, incentives, relationships, material conditions, and individual behavior interact to produce the observed outcome. It need not claim to represent the entire social system. Its purpose is to make assumptions visible, identify plausible intervention points, and show where evidence is strong, contested, or missing. The causal model then becomes the bridge between research and action because it explains why a proposed intervention might change the relevant conditions.
Research contributes most effectively when it remains connected to practical judgment. Flyvbjerg’s account of phronetic social science argues that inquiry into human affairs must address questions of values, power, direction, and desirable action rather than imitate forms of prediction associated with the natural sciences. This orientation asks where the present system is moving, who gains and loses through its operation, whether that distribution is defensible, and what practical alternatives are available (Flyvbjerg 2001).
This is not an abandonment of rigor in favor of advocacy. It is a recognition that selecting a problem, defining an outcome, and choosing among interventions already involve judgments about what matters. Research becomes more rigorous when it makes those judgments explicit and tests its interpretations against competing explanations, diverse experiences, and observable consequences. The relevant distinction is not between value-free research and value-laden action. It is between inquiry that conceals its normative assumptions and inquiry that subjects them to reasoned examination.
Treating Legal Analysis as Institutional Design
Legal analysis enters the process after research has begun, not after a preferred intervention has been fully designed. Law does more than establish an external boundary around organizational action. It distributes authority, creates rights and duties, structures procedures, defines institutional relationships, and determines which actors may make consequential decisions. Legal reasoning therefore helps translate a causal model into a legitimate institutional response.
The first legal question is one of authority. The organization must determine which institution has jurisdiction, which officials possess decision-making power, which responsibilities may be delegated, and which actions require cooperation across institutional boundaries. Many otherwise plausible interventions fail because the organization expected one actor to perform work that another actor was legally authorized, financially responsible, or procedurally equipped to undertake. Mapping legal authority clarifies where action can begin and where changes in policy, regulation, contract, or legislation may be required.
The second question concerns obligation. Statutes, regulations, constitutional principles, contracts, professional duties, and organizational policies may require certain actions while prohibiting others. A legally serious analysis considers not only whether an organization is permitted to act, but also whether it has affirmative duties toward the people affected. It examines access, nondiscrimination, confidentiality, notice, procedural fairness, professional responsibility, and the appropriate use of discretion.
The third question concerns process. Mashaw’s analysis of administrative justice demonstrates that institutional systems embody different models of legitimate decision making, including bureaucratic rationality, professional judgment, and individualized moral consideration. Each model emphasizes different values, such as consistency, expertise, responsiveness, participation, or reasoned justification, and no administrative design maximizes all of them simultaneously (Mashaw 1983).
Legal analysis must therefore move beyond a binary judgment that an intervention is either lawful or unlawful. It should ask what form of decision-making process is appropriate to the stakes, scale, and character of the problem. Highly standardized procedures may improve consistency but fail to account for relevant differences among people. Broad professional discretion may permit individualized responses but produce variation, opacity, or unequal treatment. Extensive procedural protections may strengthen legitimacy but impose burdens that make services less accessible or decisions unreasonably slow. Legal design requires balancing these values while preserving the rights and protections that the institution is obligated to honor.
Law also provides an accountability architecture. A practical intervention should specify who makes decisions, what evidence may be considered, how reasons are documented, who reviews disputed decisions, and what remedy is available when the system fails. These safeguards should be designed into ordinary workflows rather than added after harm occurs. When legal analysis is integrated early, it improves the intervention’s operational design and reduces the likelihood that implementation will depend on informal practices that cannot withstand scrutiny or leadership change.
Legal compliance remains necessary, but compliance alone is an insufficient measure of progress. A program can satisfy formal legal requirements while remaining inaccessible, arbitrary, ineffective, or ethically indefensible. Law establishes essential constraints and institutional possibilities. Ethical analysis determines how those possibilities should be used in pursuit of human and public purposes.
Making Ethical Judgment Operational
Ethical analysis translates evidence and authority into judgments about what institutions should value, whom they should prioritize, and how burdens and benefits should be distributed. Complex human problems are ethical problems because interventions inevitably advantage some interests, impose costs on others, and reflect judgments about which outcomes deserve institutional attention. Ethical reasoning should therefore be treated as part of design rather than as a final review of a completed proposal.
The first ethical task is to define the valued human outcome. Organizations often rely on administrative proxies because they are measurable, such as enrollment, case closure, service completion, or program attendance. These indicators may be useful, but they do not necessarily establish whether people have gained meaningful opportunities, improved well-being, greater agency, or more just treatment. Sen’s comparative account of justice directs attention toward actual lives, freedoms, and capabilities rather than evaluating institutions only by their formal arrangements or stated ideals (Sen 2009).
This perspective changes the evaluation question. Instead of asking only whether an institution offered a service, the organization asks whether people could realistically use it. Instead of asking only whether identical rules were applied, it asks whether relevant differences in circumstances affected people’s ability to benefit. Instead of assessing success solely through organizational outputs, it examines whether the intervention expanded substantive opportunities and reduced identifiable forms of injustice.
The second ethical task is to identify the distribution of benefits, burdens, risks, and voice. An intervention that improves average outcomes may still deepen disparities if its benefits are concentrated among people who already possess greater resources or institutional access. A program that appears efficient from an organizational perspective may shift time, transportation, documentation, or emotional burdens onto the people it is intended to assist. Ethical analysis makes these transfers visible and subjects them to justification.
The third task is to examine power. Organizations possess power to define problems, classify people, allocate resources, interpret evidence, and determine which forms of participation will influence decisions. Ethical practice requires more than inviting affected people to meetings. It requires specifying what they can shape, how disagreements will be handled, and how institutional decision makers will respond when community knowledge conflicts with professional assumptions. Practical wisdom is especially important where formal rules underdetermine the decision and where institutional actors must judge how general principles apply to particular circumstances (Flyvbjerg 2001).
The fourth task is to establish decision principles before implementation pressures intensify. These principles might include human dignity, proportionality, transparency, accessibility, reciprocity, nondiscrimination, the least restrictive effective intervention, and priority for those experiencing the greatest disadvantage. Their function is not to eliminate disagreement. They provide a shared vocabulary for explaining choices, identifying tradeoffs, and determining when a proposed efficiency would compromise the purpose of the work.
Ethical reasoning must also remain responsive to consequences. An intervention designed with benevolent intentions may create stigma, dependency, surveillance, exclusion, or unanticipated institutional burdens. Ethical evaluation should therefore continue during implementation. The relevant question is not simply whether the original design was justifiable, but whether the intervention remains justifiable in light of what participants are learning about its actual effects.
Converting Analysis into an Actionable Design
Research, legal analysis, and ethical reasoning must be synthesized into an actionable design rather than presented as three parallel reports. The synthesis should identify the desired human outcome, the causal mechanism to be influenced, the actors with authority and capacity to act, the legal rules governing their conduct, the ethical principles guiding tradeoffs, and the observable changes that would count as progress.
This synthesis can be expressed as a practical theory of change. A useful theory of change states that if particular actors undertake specified practices, under identified legal and organizational conditions, then certain mechanisms are expected to change, producing intermediate and longer-term outcomes. It should also identify the assumptions upon which this expectation depends and the risks that could prevent the intervention from functioning as intended.
The theory of change should distinguish between core functions and adaptable forms. A core function describes what the intervention must accomplish, while its form describes how that function is currently delivered. For example, the core function may be to ensure that individuals receive understandable notice and a meaningful opportunity to respond before an adverse decision. The form could involve an in-person conference, written process, digital portal, designated advocate, or combination of approaches. Preserving the function while testing different forms permits adaptation without losing legal or ethical integrity.
The resulting design should be sufficiently bounded to permit action. Complexity can produce paralysis when organizations believe they must understand and redesign the entire system before beginning. It can also produce overreach when leaders launch comprehensive reforms without the operational knowledge required to sustain them. A better approach selects an intervention point that is consequential enough to matter, limited enough to manage, and observable enough to generate learning.
Alford and Head argue that progress on complex problems requires contingent responses rather than a uniform method applied to every situation. Some problems primarily require additional knowledge, while others require negotiation among actors, changes in institutional coordination, or sustained management of conflicts that cannot be fully resolved (Alford and Head 2017).
This means that a pilot should not be designed merely as a smaller version of a predetermined program. It should function as a structured test of the causal, legal, ethical, and operational assumptions embedded in the design. The organization should know what it is trying to learn, which outcomes and process indicators will be observed, what forms of harm require immediate correction, and who has authority to modify or stop the intervention.
Treating Implementation as a Source of Knowledge
Hands-on implementation is not the final stage of translation. It is a distinct mode of inquiry that reveals whether the intervention can be enacted by real people within existing institutional conditions. Formal designs routinely underestimate workload, competing priorities, organizational culture, information limitations, resource dependencies, and the discretion exercised by frontline personnel. Implementation makes these conditions visible.
The Consolidated Framework for Implementation Research organizes implementation conditions into domains concerning the intervention, the external environment, the internal organizational setting, the people involved, and the implementation process. The framework demonstrates that the same intervention may operate differently across settings because implementation outcomes depend on the fit between the proposed practice and its institutional context (Damschroder et al. 2009).
Implementation planning should therefore examine whether participants understand the intervention, believe it serves a legitimate purpose, possess the skills and resources to perform it, and receive support from organizational leadership. It should identify how the new practice interacts with existing workflows, performance incentives, professional identities, data systems, and interorganizational dependencies. It should also determine whether the institution has created feedback mechanisms capable of detecting variation and correcting failure.
Normalization Process Theory adds that practices become embedded through continuing forms of human work. Participants must develop a shared understanding of the practice, commit themselves and others to it, perform the operational work it requires, and evaluate whether it is worthwhile and effective. These mechanisms are described as coherence, cognitive participation, collective action, and reflexive monitoring (May and Finch 2009).
This account highlights why issuing a policy or conducting a training session rarely produces durable change. People must be able to explain how the new practice differs from existing work and why the distinction matters. Relevant actors must regard participation as a legitimate part of their roles. The organization must align time, resources, authority, skills, and relationships with the demands of the practice. Participants must then receive and interpret evidence about whether the practice is producing sufficient value to justify continued effort (May and Finch 2009).
Hands-on implementation also reveals conflicts among research findings, legal requirements, ethical commitments, and institutional capacity. A legally protective process may prove too difficult for intended participants to navigate. An evidence-based intervention may rely on professional roles that the organization cannot staff consistently. A technically efficient procedure may undermine trust or impose unequal burdens. These tensions should not be treated as reasons to abandon interdisciplinary analysis. They are precisely the knowledge that integrated implementation is designed to produce.
The response should be structured adaptation. Researchers reconsider the causal model, legal experts determine whether alternative procedures preserve required protections, ethical analysis examines how modifications affect people and values, and practitioners test revised forms. Each cycle should preserve the intervention’s essential purpose while improving its fit with the environment.
Measuring Progress as Learning and Institutional Change
Complexity should alter how progress is measured. A single outcome measure can conceal whether an intervention reached the intended population, operated through the expected mechanism, distributed benefits fairly, complied with legal requirements, or became embedded in ordinary practice. Evaluation should therefore examine outcomes, implementation, legality, equity, experience, and institutional capacity.
Outcome measures assess whether the relevant human condition changed. Implementation measures determine whether the practices believed to produce that change actually occurred. Legal measures examine consistency, timeliness, reasons, procedural protections, and access to review. Ethical measures examine distribution, participation, burden, dignity, and the experiences of those affected. Capacity measures determine whether the organization has developed the roles, relationships, knowledge, routines, and resources necessary to continue improving the system.
These measures should be disaggregated whenever aggregate results might conceal meaningful differences among groups or settings. They should also be interpreted collectively. Quantitative patterns can identify where outcomes vary, while qualitative inquiry can help determine whether those variations reflect context, implementation failure, institutional barriers, or a mistaken causal assumption. The purpose of measurement is not simply to demonstrate that the intervention worked. It is to determine what worked, for whom, under what conditions, through which mechanisms, and at what human or institutional cost.
Progress may include changes that precede the ultimate outcome. These can include greater clarity about the problem, improved coordination among institutions, more transparent decision processes, reduced procedural burden, strengthened trust, better information, increased community influence, or improved capacity for collective action. Such changes should not be used to avoid accountability for substantive results. They are important when they plausibly increase the system’s capacity to produce and sustain those results.
For deeply complex problems, progress may also involve preventing deterioration, reducing the severity of harm, or improving the institution’s ability to respond adaptively. Alford and Head reject the expectation that every wicked problem can be conclusively solved and propose a more realistic emphasis on improvement and better management. This standard does not lower expectations so much as locate accountability in observable movement, responsible learning, and improved capacity rather than rhetorical claims of resolution (Alford and Head 2017).
Aligning People, Purpose, and Practice
Translating complexity into progress ultimately requires alignment across people, purpose, and practice. People must include those who possess relevant knowledge, legal authority, implementation responsibility, and lived experience. Their roles should be differentiated but connected, allowing each form of expertise to challenge the assumptions of the others.
Purpose must define the human outcome, public value, and institutional responsibility that justify the work. It should function as a criterion for deciding which evidence matters, which legal authorities should be exercised, which ethical principles should govern tradeoffs, and which adaptations remain faithful to the intervention’s reason for existing.
Practice must convert purpose into routines that people can actually perform. It includes decision procedures, communication channels, documentation, training, allocation of responsibility, feedback, evaluation, and revision. A practice is not aligned merely because it is described in a strategic plan. It is aligned when the people responsible for it understand its purpose, possess the authority and capacity to carry it out, and can evaluate whether it is advancing the intended outcome.
The relationship among these elements is recursive. Research may reveal that the organization has defined the wrong population or mechanism. Legal analysis may show that authority lies elsewhere or that existing procedures inadequately protect affected people. Ethical reasoning may expose an inequitable distribution of burdens. Implementation may demonstrate that a theoretically sound practice cannot be sustained in its current form. Each finding requires renewed alignment rather than defense of the original design.
The distinctive contribution of an integrative approach is not that it eliminates complexity. It makes complexity governable. Research converts uncertainty into a credible account of the problem. Legal analysis converts authority and obligation into an institutional architecture. Ethical reasoning converts competing values into defensible principles and judgments. Implementation converts proposals into situated practices and generates new knowledge about their consequences. Together, these disciplines create a cycle through which organizations can move from observation to judgment, from judgment to action, and from action to learning.
Progress occurs when this cycle produces more than an isolated intervention. It should leave behind a stronger capacity to understand problems, justify decisions, coordinate authority, protect people, implement practices, and revise systems in light of evidence. The immediate objective may be to improve a program, policy, process, or partnership. The deeper objective is to develop institutions capable of learning responsibly while acting under conditions of uncertainty.
Brandon L. Blankenship is an assistant professor, continuing legal education presenter, and business educator. He and his wife Donnalee live on their hobby farm south of Birmingham, Alabama.
Alford, John, and Brian W. Head. 2017. “Wicked and Less Wicked Problems: A Typology and a Contingency Framework.” Policy and Society 36 (3): 397–413.
Damschroder, Laura J., David C. Aron, Rosalind E. Keith, Susan R. Kirsh, Jeffery A. Alexander, and Julie C. Lowery. 2009. “Fostering Implementation of Health Services Research Findings into Practice: A Consolidated Framework for Advancing Implementation Science.” Implementation Science 4: 50.
Flyvbjerg, Bent. 2001. Making Social Science Matter: Why Social Inquiry Fails and How It Can Succeed Again. Cambridge: Cambridge University Press.
Mashaw, Jerry L. 1983. Bureaucratic Justice: Managing Social Security Disability Claims. New Haven, CT: Yale University Press.
May, Carl, and Tracy Finch. 2009. “Implementing, Embedding, and Integrating Practices: An Outline of Normalization Process Theory.” Sociology 43 (3): 535–554.
Sen, Amartya. 2009. The Idea of Justice. Cambridge, MA: Harvard University Press.
Van de Ven, Andrew H. 2007. Engaged Scholarship: A Guide for Organizational and Social Research. Oxford: Oxford University Press.
Designing practical systems and partnerships that last requires a different orientation from launching a successful project. A project is organized around producing a defined result within a specified period, while a durable system must continue coordinating people, resources, decisions, and practices after the initial urgency, funding, or leadership attention has diminished. Sustainability should therefore not be defined as preserving an intervention exactly as it was first designed. It is the capacity to preserve a valued purpose, maintain the practices necessary to advance it, and adapt those practices as organizational and environmental conditions change (Chambers, Glasgow, and Stange 2013; Stirman et al. 2012).
Within a framework of aligning people, purpose, and practice, durable design begins by treating complex human problems as systems of interdependence rather than isolated deficiencies. Problems such as community violence, educational inequality, public health disparities, housing instability, and limited access to justice are produced and sustained through interactions among institutions, policies, professional practices, material conditions, and human relationships. No single organization usually possesses the authority, knowledge, legitimacy, or operational capacity required to address every dimension of such a problem. Cross-sector partnerships become necessary when relevant resources and responsibilities are distributed across governmental agencies, nonprofit organizations, businesses, universities, community groups, and affected populations (Bryson, Crosby, and Stone 2015; Emerson, Nabatchi, and Balogh 2012).
The first design task is to establish a durable purpose. Partnerships are often formed around an available grant, a charismatic leader, a public crisis, or an attractive program model. These conditions can mobilize participation, but they provide a weak foundation for long-term cooperation if the partners have not developed a shared understanding of the problem they are trying to solve. The partnership must articulate the population or community it exists to serve, the conditions it seeks to change, the public value it intends to create, and the limits of what it can accomplish. This shared purpose should be sufficiently specific to guide resource allocation and accountability, yet sufficiently broad to permit different institutions to contribute according to their distinctive competencies (Bryson, Crosby, and Stone 2015).
A durable purpose is not merely a mission statement. It functions as a decision rule. When opportunities, funding requests, leadership changes, and institutional pressures arise, the partnership should be able to ask whether a proposed activity advances the shared purpose or simply adds another disconnected initiative. Purpose establishes continuity across changing projects because individual programs can be modified, replaced, or discontinued while the underlying commitment remains stable. The relevant question is therefore not whether a particular activity continues indefinitely, but whether the system retains the capacity to produce the outcomes for which the activity was originally created (Stirman et al. 2012).
Shared purpose must be accompanied by a common theory of change. Partners should identify the conditions producing the problem, the mechanisms through which their collective activities are expected to change those conditions, and the intermediate outcomes that would indicate progress. A theory of change enables participants to test whether their activities are connected to their intended results rather than assuming that collaboration itself constitutes success. It also helps distinguish between core functions that must be protected and delivery methods that can be adapted. Without this distinction, organizations may preserve visible program components after those components have ceased to serve their original purpose (Chambers, Glasgow, and Stange 2013).
Durability also depends on aligning the partnership’s governance structure with the complexity of the work. Networks can be governed collectively by participating organizations, coordinated by a dominant lead organization, or administered through a separate network-level entity. Each form creates different advantages and vulnerabilities, and no single governance arrangement is universally appropriate. Shared governance can strengthen ownership and legitimacy, but it can become inefficient as the number of partners and the need for coordination increase. Lead-organization governance can provide direction and administrative capacity, but it can also concentrate authority and weaken the commitment of other participants. A separate network administrative organization can create stability and professional coordination, although it requires resources and may distance governance from frontline participants (Provan and Kenis 2008).
The governance model should be selected according to the number of participants, the level of trust among them, the degree of agreement about goals, and the need for centralized coordination. A small partnership with high trust and a narrowly defined task may function through shared governance, while a large and diverse network responsible for complex service delivery may require a dedicated coordinating body. As the partnership develops, its governance arrangements should be reviewed rather than treated as permanent. A structure appropriate during formation may become inadequate when the partnership expands, assumes new responsibilities, or moves from planning to implementation (Provan and Kenis 2008).
Whatever governance form is selected, durable partnerships require explicit decision rights. Participants should know who may set priorities, commit resources, modify practices, speak on behalf of the partnership, admit new members, and resolve disputes. Ambiguity can appear collaborative because no organization seems to dominate, but unclear authority often delays action and permits participants to avoid responsibility. Durable governance distributes influence without dissolving accountability. It identifies which decisions require consensus, which require a majority or supermajority, which may be delegated to staff, and which remain within the authority of individual partner organizations (Bryson, Crosby, and Stone 2015).
Formal agreements have an important role in this architecture, but written agreements alone do not create a functioning partnership. Memoranda of understanding, contracts, bylaws, data-sharing agreements, meeting schedules, and reporting requirements can clarify expectations and protect continuity. Their practical value depends on whether they reinforce the relationships through which people exchange information, solve problems, and fulfill commitments. Research on collaboration within local governments indicates that informal communication and voluntary interaction contribute directly to trust and collaborative behavior, while formal mechanisms can strengthen collaboration indirectly by creating the conditions in which those informal relationships develop and continue (Park, Krause, and Hawkins 2021).
Sustainable design therefore requires both relational and structural infrastructure. Relational infrastructure consists of trust, reciprocity, mutual understanding, credible commitments, and the ability to address disagreement without threatening the partnership. Structural infrastructure consists of roles, procedures, meeting routines, accountability mechanisms, information systems, and formal authority. Relationships without structure can become dependent on personal goodwill, while structure without relationships can produce compliance without meaningful cooperation. Durable systems emerge when formal arrangements support repeated human interaction and those interactions, in turn, give institutional arrangements practical legitimacy (Park, Krause, and Hawkins 2021).
Meaningful ownership is another condition of durability. People are more likely to maintain systems when they participate in shaping the rules that govern their work and when those rules correspond to the realities they encounter. Ostrom’s analysis of long-enduring institutions demonstrates that durable collective arrangements frequently include clear boundaries, locally appropriate rules, participant involvement in rule modification, monitoring, graduated responses to noncompliance, accessible conflict-resolution processes, and governance structures nested across multiple levels (Ostrom 1990).
These principles suggest that sustainability cannot be imposed entirely from above. Executive sponsorship may authorize collaboration, provide resources, and remove institutional barriers, but frontline practitioners and affected communities must be able to influence operational design. They know where referral systems break down, which procedures discourage participation, how institutional requirements conflict, and what forms of assistance are usable in daily life. Their knowledge is not supplementary to system design. It is necessary for ensuring that formal arrangements correspond to the conditions under which implementation actually occurs (Ostrom 1990; Emerson, Nabatchi, and Balogh 2012).
Participation should nevertheless be structured rather than romanticized. Not every participant needs authority over every decision, and broad inclusion without differentiated responsibilities may produce confusion. Durable partnerships specify who is affected, who possesses relevant expertise, who controls essential resources, who implements the work, and who has authority to change institutional conditions. They then create appropriate forms of participation for each group. A resident may possess indispensable experiential knowledge without controlling an agency budget, while an agency executive may control resources without understanding how a policy operates at the point of service. Sustainable design integrates these forms of knowledge and authority without pretending that they are identical (Bryson, Crosby, and Stone 2015).
The system must also convert commitments into repeatable practice. Partnerships become durable when their essential activities are embedded in organizational routines rather than performed as extraordinary acts by unusually committed individuals. Referral processes, training protocols, data collection, joint case review, community consultation, resource allocation, and evaluation should be assigned to identifiable roles and incorporated into normal workflows. When collaborative activity remains outside participants’ formal responsibilities, it is especially vulnerable to workload pressures, staff turnover, and leadership transitions (Stirman et al. 2012).
Institutionalization does not require creating a large bureaucracy. It requires making the work visible, assignable, teachable, and transferable. Each critical function should have an owner, a documented process, a defined schedule, and a means of verifying completion. Knowledge should reside in shared records and practices rather than exclusively in the memory or relationships of founding participants. New personnel should be able to understand why the partnership exists, what commitments have been made, how decisions are reached, and what they are expected to contribute. Succession planning is therefore not limited to replacing senior leaders. It includes reproducing the knowledge, relationships, and practical competencies that allow the system to operate (Bryson, Crosby, and Stone 2015).
Financial design must likewise extend beyond obtaining start-up funding. Temporary grants can support experimentation, but programs become fragile when the activities that produce value remain permanently dependent on discretionary or short-term funds. Partners should determine which functions require recurring financial support, which can be integrated into existing institutional responsibilities, which produce value for participating organizations, and which may require diversified revenue. Sustainability is strengthened when essential practices are included in operating budgets, personnel allocations, service contracts, reimbursement mechanisms, or other recurring institutional commitments (Stirman et al. 2012).
Partners should also examine the distribution of costs and benefits. Collaborative arrangements can deteriorate when one organization supplies most of the labor while others receive reputational, financial, or operational benefits. Contributions do not need to be equal, but they should be transparent and perceived as appropriate to each participant’s capacity and benefit. Agreements should identify contributions of money, personnel, facilities, data, authority, expertise, and community access. This makes hidden subsidies visible and enables the partnership to address imbalances before they become sources of resentment or withdrawal (Bryson, Crosby, and Stone 2015).
Measurement is another component of durable practice. A sustainable partnership needs evidence that its activities are producing value, but its evaluation system should distinguish among implementation, outcomes, relational health, and institutional capacity. Implementation measures determine whether agreed practices are occurring. Outcome measures assess whether conditions are changing for the people or communities served. Relational measures examine trust, participation, information exchange, and fulfillment of commitments. Capacity measures assess whether the system has the personnel, resources, authority, and routines required to continue the work (Emerson, Nabatchi, and Balogh 2012; Stirman et al. 2012).
These measures should support learning rather than merely satisfy reporting requirements. Data become useful when partners review them together, examine variation, identify unintended consequences, and modify their practices. An evaluation system that produces annual reports without affecting decisions may document activity while contributing little to sustainability. Durable systems shorten the distance between evidence and action by establishing regular occasions for interpreting results and authorizing changes (Chambers, Glasgow, and Stange 2013).
This learning orientation is crucial because sustainability does not mean resistance to change. The Dynamic Sustainability Framework argues that interventions, organizational settings, and broader environments continue changing after implementation. Effective sustainment therefore depends on repeatedly examining the fit among the intervention, the practice setting, and the wider system. Adaptation becomes a normal feature of sustainability rather than evidence that the original program has failed (Chambers, Glasgow, and Stange 2013).
The practical challenge is to distinguish responsible adaptation from loss of purpose. Partnerships should identify core principles and functions that must remain stable, while permitting peripheral procedures to change. For example, the core function may be ensuring that people receive coordinated access to legal, health, and social services, while the specific referral technology, meeting schedule, or staffing arrangement may evolve. Preserving a procedure after it no longer fits the context can undermine sustainability just as surely as abandoning the system altogether. Durability depends on continuity of function combined with flexibility of form (Chambers, Glasgow, and Stange 2013).
Periodic redesign should therefore be built into the partnership’s governance cycle. Partners might conduct scheduled reviews of their purpose, membership, governance, resource model, practices, and outcomes. These reviews should ask whether the original problem has changed, whether the appropriate people remain involved, whether authority matches responsibility, whether benefits and burdens are fairly distributed, and whether existing routines continue producing value. By authorizing revision in advance, the partnership reduces the likelihood that adaptation will be interpreted as institutional failure or political disloyalty (Ostrom 1990; Chambers, Glasgow, and Stange 2013).
Durable partnerships must finally be designed to survive the departure of their founders. Charismatic leadership can generate initial commitment, bridge institutional divisions, and attract resources. It can also create dependency when relationships, knowledge, and authority remain concentrated in one person. Sustainable leadership is distributed across multiple levels, including executive sponsors who provide legitimacy, boundary spanners who connect institutions, managers who coordinate operations, frontline practitioners who adapt the work, and community participants who assess whether the system remains responsive. This distribution creates redundancy, protects continuity, and prevents the partnership from being equated with a single leader’s tenure (Emerson, Nabatchi, and Balogh 2012; Provan and Kenis 2008).
Designing practical systems and partnerships that last is therefore a process of institutionalizing alignment. People are aligned by giving relevant actors meaningful roles, distributing authority, cultivating trust, and making contributions visible. Purpose is aligned by defining the shared challenge, articulating the public value to be created, and establishing a theory of how collective action will produce change. Practice is aligned by embedding responsibilities in routines, financing essential functions, measuring outcomes, preserving core principles, and adapting delivery methods as conditions evolve. The goal is not to build an arrangement that never changes. It is to build one that can continue learning, coordinating, and producing value without depending on a temporary crisis, a single grant, or an indispensable individual.
A quick fix attempts to solve the visible problem in front of the organization. A durable system develops the collective capacity to recognize, interpret, and respond to the conditions that repeatedly produce the problem. Sustainable partnerships last because they are governed, resourced, practiced, evaluated, and renewed. Their endurance is not passive persistence. It is the disciplined capacity to preserve purpose through responsible change.
Brandon L. Blankenship is an assistant professor, continuing legal education presenter, and business educator. He and his wife Donnalee live on their hobby farm south of Birmingham, Alabama.
Bryson, John M., Barbara C. Crosby, and Melissa Middleton Stone. 2015. “Designing and Implementing Cross-Sector Collaborations: Needed and Challenging.” Public Administration Review 75 (5): 647–663.
Chambers, David A., Russell E. Glasgow, and Kurt C. Stange. 2013. “The Dynamic Sustainability Framework: Addressing the Paradox of Sustainment amid Ongoing Change.” Implementation Science 8: 117.
Emerson, Kirk, Tina Nabatchi, and Stephen Balogh. 2012. “An Integrative Framework for Collaborative Governance.” Journal of Public Administration Research and Theory 22 (1): 1–29.
Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press.
Park, Angela Y. S., Rachel M. Krause, and Christopher V. Hawkins. 2021. “Institutional Mechanisms for Local Sustainability Collaboration: Assessing the Duality of Formal and Informal Mechanisms in Promoting Collaborative Processes.” Journal of Public Administration Research and Theory 31 (2): 434–450.
Provan, Keith G., and Patrick Kenis. 2008. “Modes of Network Governance: Structure, Management, and Effectiveness.” Journal of Public Administration Research and Theory 18 (2): 229–252.
Stirman, Shannon Wiltsey, John Kimberly, Natasha Cook, Amber Calloway, Frank Castro, and Martin Charns. 2012. “The Sustainability of New Programs and Innovations: A Review of the Empirical Literature and Recommendations for Future Research.” Implementation Science 7: 17.
Within an approach centered on aligning people, purpose, and practice, bringing people together should be understood as more than convening representatives or facilitating productive meetings. It is the deliberate construction of a collaborative system through which participants can understand their interdependence, combine distinct forms of knowledge, and exercise coordinated agency. Complex human problems routinely cross professional, organizational, and jurisdictional boundaries, while the knowledge, authority, legitimacy, and resources needed to address them remain distributed among multiple actors. Collaboration becomes necessary when no institution possesses either a sufficient understanding of the problem or the capacity to resolve it independently (Bryson, Crosby, and Stone 2006; Emerson, Nabatchi, and Balogh 2012).
The first task is to identify the people who constitute the relevant problem system. Participants should be selected not simply because they hold senior positions, but because they possess some combination of disciplinary knowledge, institutional authority, operational capacity, community legitimacy, implementation responsibility, or direct experience of the problem. A stakeholder map can clarify who experiences the consequences of the challenge, who influences the conditions producing it, who controls relevant resources, and who will ultimately carry out any proposed response. This approach broadens participation beyond conventional institutional representatives while preserving a clear rationale for why each person is at the table. Research on collaborative governance indicates that participation is shaped by incentives, prior histories of conflict or cooperation, resource inequalities, and participants’ expectations that their involvement will matter (Ansell and Gash 2008).
Representation alone, however, does not guarantee meaningful inclusion. Collaborative processes may reproduce existing hierarchies when well-resourced institutions determine the agenda, define the acceptable evidence, or control which proposals are treated as feasible. Participation must therefore be consequential rather than ceremonial. Conveners should establish transparent criteria for membership, provide participants with comparable access to information, create opportunities for less powerful actors to shape the agenda, and clarify how recommendations will affect institutional decisions. The quality of deliberation is particularly important because respectful consideration of opposing viewpoints can improve the influence of minority perspectives even when those perspectives are numerically underrepresented (Kim 2024).
Once the relevant people have been assembled, the group must develop a shared purpose. Shared purpose should not be confused with a predetermined mission statement presented by the convener. It is better understood as a negotiated working hypothesis about the problem, the public value at stake, and the change the participants are prepared to pursue together. Participants may disagree about causes, priorities, or preferred interventions while still reaching a sufficiently shared understanding of what requires collective attention. Collaborative-governance research identifies principled engagement, shared motivation, and joint capacity as mutually reinforcing dimensions of effective collaboration, which means that purpose develops through interaction rather than being settled before interaction begins (Emerson, Nabatchi, and Balogh 2012).
A disciplined problem-framing process should help participants distinguish symptoms from underlying conditions, identify who is affected, examine how different institutions contribute to or constrain the problem, and specify what success would look like from multiple perspectives. The resulting purpose must be broad enough to accommodate different forms of contribution but precise enough to guide decisions and evaluate progress. It should also identify the limits of the collaboration, including which issues fall outside its authority and which outcomes require action from external institutions. Face-to-face dialogue, trust building, shared understanding, and visible early achievements can create a reinforcing cycle in which participants become increasingly willing to invest in the collaborative process (Ansell and Gash 2008).
Interdisciplinary and interinstitutional collaboration also requires active management of knowledge boundaries. Participants from different fields may use the same words differently, recognize different forms of evidence, and assess proposed actions according to incompatible professional standards. These boundaries become progressively more difficult as collaboration moves from exchanging information to interpreting meanings and then to changing established practices or interests. Knowledge can be transferred when participants share a common vocabulary, translated when they must reconcile different interpretations, and transformed when collaboration requires them to alter existing assumptions, priorities, or routines (Carlile 2004).
Effective convening therefore creates boundary objects that allow participants to work on the same problem without erasing their disciplinary differences. A jointly developed problem statement, system map, case narrative, theory of change, shared dataset, prototype, or performance dashboard can provide a common point of reference. Such tools should be concrete enough to organize joint work but adaptable enough to remain intelligible within several professional and institutional contexts. The objective is not to make every participant think alike. It is to create sufficient common knowledge for participants to understand where their perspectives converge, where they conflict, and what those differences mean for collective action (Carlile 2004).
This work frequently depends on a boundary spanner who can move among institutional environments while retaining credibility within each of them. Boundary spanners build networks, interpret professional languages, coordinate activity, broker relationships, and identify opportunities for joint action. These functions have been described as those of the reticulist, interpreter, coordinator, and entrepreneur, with each role requiring a distinct combination of relational skill, institutional knowledge, and practical judgment (Williams 2012).
The boundary spanner is not merely a neutral facilitator. The role requires enough independence to ask difficult questions, enough institutional access to mobilize resources, and enough relational legitimacy to remain trusted when disagreements arise. This position also involves political judgment because decisions about who participates, which evidence is recognized, and what counts as progress inevitably affect the distribution of influence. A boundary spanner can make collaboration possible, but sustainable collaboration cannot depend entirely on one charismatic individual. Institutions must authorize the work, allocate resources, protect the time required for participation, and embed collaborative responsibilities within organizational structures (Bryson, Crosby, and Stone 2006; Williams 2012).
Alignment around practice converts shared purpose into coordinated action. Collaborators need an explicit governance architecture that identifies roles, decision rules, resource commitments, communication channels, accountability mechanisms, and procedures for addressing conflict. Without these arrangements, collaboration can remain an exchange of ideas rather than become a means of implementation. Effective cross-sector collaborations adapt their structures to the problem, the participating organizations, and the institutional environment rather than imposing a single governance model across every context (Bryson, Crosby, and Stone 2006).
The work should proceed through iterative cycles of diagnosis, design, action, reflection, and adaptation. Early projects should be substantial enough to demonstrate the value of collaboration but bounded enough to permit learning and correction. Small wins can strengthen trust and commitment, particularly where participants have limited experience working together or inherit a history of conflict (Ansell and Gash 2008). These early achievements should not become substitutes for systemic change. They should function as tests of the collaboration’s assumptions, opportunities to strengthen joint capacity, and foundations for more consequential institutional action.
Sustaining engagement requires more than initial enthusiasm. An empirical study of participation in an environmental justice council found that repeated positive relationships were associated with increased attendance, while repeated negative relationships were associated with decreased attendance. One-time interactions did not produce the same associations, which suggests that durable engagement develops through accumulated relational experience rather than isolated convenings (Ambrose and Siddiki 2024). The same study indicates that sustained participation is not synonymous with uniform agreement because participants may remain engaged around oppositional or unresolved objectives when the forum continues to provide a meaningful arena for influence (Ambrose and Siddiki 2024).
Collaborative leadership must therefore protect both relationship and disagreement. Consensus should remain an aspiration for coordinated action, but it should not be used to suppress minority viewpoints, conceal unequal power, or prematurely close debate. High-quality deliberation requires participants to explain their reasoning, respond to contrary evidence, and demonstrate how minority concerns were considered in the final decision. Structured dissent, transparent documentation of unresolved differences, and publicly stated rationales for decisions can strengthen both the substantive quality and perceived fairness of collaborative outcomes (Kim 2024).
Bringing people together across disciplines and institutions is thus a process of building a temporary but disciplined collaborative system. People are aligned through legitimate inclusion, relationship building, and attention to power. Purpose is aligned through joint problem framing, negotiated public value, and shared criteria for success. Practice is aligned through translation, governance arrangements, iterative action, and mutual accountability. The result is not consensus for its own sake. It is the capacity of diverse people and institutions to understand a shared challenge more fully, make better decisions together, and coordinate action that none could accomplish alone.
Brandon L. Blankenship is an assistant professor, continuing legal education presenter, and business educator. He and his wife Donnalee live on their hobby farm south of Birmingham, Alabama.
Ambrose, Graham, and Saba Siddiki. 2024. “Assessing Drivers of Sustained Engagement in Collaborative Governance Arrangements.” Journal of Public Administration Research and Theory 34 (4): 498–514.
Ansell, Chris, and Alison Gash. 2008. “Collaborative Governance in Theory and Practice.” Journal of Public Administration Research and Theory 18 (4): 543–571.
Bryson, John M., Barbara C. Crosby, and Melissa Middleton Stone. 2006. “The Design and Implementation of Cross-Sector Collaborations: Propositions from the Literature.” Public Administration Review 66 (S1): 44–55.
Carlile, Paul R. 2004. “Transferring, Translating, and Transforming: An Integrative Framework for Managing Knowledge Across Boundaries.” Organization Science 15 (5): 555–568.
Emerson, Kirk, Tina Nabatchi, and Stephen Balogh. 2012. “An Integrative Framework for Collaborative Governance.” Journal of Public Administration Research and Theory 22 (1): 1–29.
Kim, Jiho. 2024. “Distributive Justice in Collaborative Outputs: Empowering Minority Viewpoints Through Deliberation.” Journal of Public Administration Research and Theory 34 (1): 1–15.
Williams, Paul. 2012. Collaboration in Public Policy and Practice: Perspectives on Boundary Spanners. Bristol: Policy Press.
