What Faiths Demand of the State

A faith that says anything about conduct ends up with a position on law, because law is where conduct is enforced. Those positions are presented as consequences of the faith rather than as policy preferences. That is what makes them examinable. A demand that claims to follow from a doctrine either follows from it or does not, and someone who holds no view on whether the doctrine is true can settle that question.

The premises used are the tradition’s own commitments: the text is authoritative, the doctrine binds, and the authorities named inside it have the standing claimed for them. Granting all of that is what makes the inference visible. Whether a faith is true is a separate question, and granting the premises leaves that question open. The defenses that bear only on that question are set aside: textual preservation, fulfilled prophecy, scientific content, the founder’s conduct, and literary quality. Two defenses of divine standing are kept, and they are taken up for what they license about a political order rather than for what they establish about a creed.

The demands divide into two kinds. Some are demands on policy: what the state must prohibit, permit, or fund. Others are demands on the form of the polity: who counts as a member, and what holds the members together. The two kinds fail in different places.

Core takeaways

• A demand a faith makes on the state can be checked by granting the faith’s own premises and asking whether the demand follows. The check does not require any view on whether the faith is true.
• The demands that survive the check state a moral constraint and stop there.
• The demands that fail are the ones that carry a specific institution out of a doctrinal premise. Each of those needs a second premise about institutions, and that premise is usually missing.
• Two demands in this material conflict with each other. The just-war condition requires a forecast of the harm a remedy causes, and the argument from the scale of abortion suspends that forecast.
• The argument that a polity needs a religion works when religion means a value ordering encoded in law. The conclusion it draws needs religion to mean a confessional faith.
• The pattern is not peculiar to religious argument. It is easier to see because the first premise is written down in a text.

Abortion

Half of the position is the tradition’s own premise and is granted rather than argued: what is killed is a human being. The case for permission has largely moved to bodily autonomy, which concedes what the act is and argues past it. Granting the premise, the demand is short. A state that prohibits the killing of innocents prohibits this, and it has no more discretion here than for any other homicide.

The contested part is what to do about it. Two camps accept the verdict and divide on method. One supports any law that reduces the number, even where the law stops short of prohibition. The other refuses partial measures, on the ground that a partial prohibition legislates the permitted remainder. Both camps describe the dispute as one about strategy.

A third argument enters the dispute from scale. The running total is in the billions and the rate is rising, which makes this the largest act of its kind humans have carried out. On that argument, whatever is not working must be dropped now, and whatever can work must be taken up now, however unthinkable it was before.

The scale premise is granted. The question is what it licenses. An urgency claim can establish that the present approach is inadequate to the size of the problem. It cannot establish which approach is adequate, because that is a claim about what would actually reduce the count, and no doctrine contains that information. “Whatever can work” also states no upper bound. With no bound stated, it approves in advance any action its proposer believes will work. Anyone who has assigned a large enough number to a wrong can use the same sentence.

The argument also settles the strategy dispute by counting outcomes, which is the first camp’s method. Put to the second camp, whose position is that principle is not decided by counts, it restates the view they had already declined. It also conflicts with the conditions the same case sets for taking action against a government.

When resistance to a government is allowed

Render to Caesar what is Caesar’s and to God what is God’s. The formula divides two claims: the state receives what belongs to it, and what belongs to God is not the state’s to dispose of. Obedience is therefore conditional. The passages instructing submission to governing authority cannot be read as unconditional, because that reading removes the division those passages rest on.

The conditions worked out for a war being just are three. There must be a valid cause: self-defense, recovery of something unjustly taken, or the correction of a grave wrong. There must be right intention, with a durable peace as the aim. And the action must proceed from a legitimate authority. The third condition rules out a private individual acting alone. For a tyranny the tradition supplies a worked-out route around that condition. Tyranny counts as violence rather than as law, so resistance to it is not sedition against law, and legitimacy in that case rests with lesser officials, who are themselves bound to the higher law, together with the community as a body. Two further conditions apply. The overthrow must not produce an evil greater than the tyranny it removes, and the legal channels must have been exhausted first.

As a set of rules this works, and it is more restrictive than any other demand in either tradition. It authorizes neither a private individual nor a popular majority. It makes the cost of the remedy a condition to be met rather than a factor to be weighed against how badly the remedy is wanted.

The strain is in application. Three of the five conditions are assessed by whoever proposes the action: cause, intention, and the comparison of evils. The comparison is a forecast about a war that has not happened, which is the one thing nobody can check in advance. The condition checkable from outside is the one about authority, and it places the decision with officeholders. The framework therefore binds to exactly the degree that officials exist who treat it as binding. That is a real constraint in a polity that has such officials and no constraint at all in a polity that does not.

A second problem is a conflict between two demands. The greater-evil condition and the argument from scale cannot both be operating. One requires a forecast of the harm the remedy causes and refuses the remedy when the forecast is bad. The other suspends the forecast on the ground that the harm being addressed is too large to wait on it. A case carrying both leaves the choice of which one governs to whoever is holding the case at the time.

Charity and the state

The obligation to give is not in dispute. The criterion for receiving is. Need is set by inability rather than by want. Whoever can provide for himself and does not falls outside the class the obligation covers. The rule extends to those with families able to support them, and to women of marriageable age, because a means of provision is available in both cases. The reason is arithmetic about a finite pool. A recipient who could have provided for himself has consumed the share of one who could not. The test for need exists to protect the desperate rather than to spare the donor.

The institutional conclusion follows from there. The state has proven bad at administering the test and prone to corruption in the attempt. It also works under an incentive that guarantees the test erodes, because campaigns are won by promising transfers to people who could work, which is the purchase of votes with money taken from others. The conclusion drawn is that the obligation is discharged personally and through the community, and that state welfare in its current forms should be opposed.

Those are two steps, and the first is much stronger than the second. The criterion is a moral claim and it rests on the text. The extension to a means not yet taken up counts a possible future provider as a present ability, which is a judgment about availability rather than a reading of the rule. The conclusion is a claim about institutions, and it needs a further premise: that no state can administer a criterion of this kind. What is supplied instead is that this state administers it badly and corruptly. That is a claim about one record, and the conclusion requires a claim about states in general.

The alternative has the same problem it was offered to solve. Somebody still has to determine ability. A community almoner makes the same assessment, with less information than he is usually credited with, and under the same incentive to be generous with funds he did not earn. The scale is smaller and the structure is identical. The appeal to how charity was handled in the past establishes that the arrangement existed. It does not establish that the arrangement met a comparable volume of need.

The argument establishes that a welfare system designed without the test violates the criterion. That result holds, and it is narrower than the conclusion drawn from it.

The foreigner living among a people

The text repeatedly requires that the foreigner living among a people be equal under its law. Western practice is not equality. It gives preferential access to transfer programs ahead of natives, softer handling by police, and effective immunity in certain categories of offense. Whatever those passages command, the current arrangement is not it. The reading that turns them into a command to admit without limit replaces one proposition with another, because a rule about how the foreigner already present is treated says nothing about how many are admitted.

That much holds without qualification. The step from a rule of treatment to a rule of admission is not available, and the passages are cited for admission far more often than they license.

The rule constrains in both directions, which the case using it does not keep in view. Equality under the law forbids the preference just described, and it forbids a penalty imposed the other way. The text settles the treatment of those present. It leaves the number admitted undetermined.

The second argument concerns what a nation is. The text names peoples by descent, identifying a people as the descendants of a common ancestor, and it warns about the consequences of importing foreign cultures. This is taken to support the position that a nation is its people rather than its bureaucracy, so that replacing the people replaces the nation.

The premise is that the text uses descent to identify peoples, which establishes a usage. The conclusion is that a polity ought to be constituted by descent, which is a rule for action. A text can name groups by ancestry from beginning to end and still issue no instruction about the composition of a state. No step in the argument crosses from the naming convention to the rule. The nation-is-its-people position may be defensible on other grounds. It is not defensible from a naming convention.

A further observation is made about who arrives: a population claiming refuge is disproportionately men of military age, and it does not return once the wars end. That is an argument about whether a legal category has been applied correctly. It bears on classification and administration. It does not bear on what the text requires toward those inside the category, which is the same whether they were placed there rightly or wrongly.

Homosexual conduct and the penalty that was dropped

The older civil code prescribed a severe penalty for homosexual conduct. That code was given to a particular people as their civil law and does not bind Christians now. The people making the argument concede this at the start rather than under pressure. What remains binding is the moral verdict on the conduct. The argument marks the difference between a code that penalized the conduct and an order that legalizes it and actively promotes it. Loving the sinner is a real obligation, and it does not extend to excusing the conduct or suspending justice about it.

The case also concedes a problem with its reach. Most of the people making it can say only that the text prohibits the conduct. That is true, and it settles the matter for anyone who already accepts the text. To everyone else it reads as plain hostility. A reason a non-believer could check is owed, and it is generally not supplied.

That concession costs more than it is usually taken to cost. Once the civil penalty is disclaimed, the content of the demand on the state is unspecified. The argument establishes a verdict on the conduct and a prohibition on the state endorsing it. It does not establish a penalty, because the penalty the code contained has been set aside. It does not fix where the state should sit between prohibiting the conduct and promoting it. Everything between those two points is what is actually in dispute, and the argument supplies no rule for picking a point.

Whether democracy is required

No passage commands any particular form of government. The obligation is to judge wisely which arrangement suits the society in question, and an arrangement that suits one society can fail another. Treating a system as an absolute good, so that any threat to it counts as sacrilege, gives a political arrangement a standing the faith reserves for something else.

This is granted in full. For a tradition whose political obligations come from its text, the absence of a command is the absence of an obligation, and the argument claims nothing beyond that. The point about idolatry is a substantive claim. A commitment held as beyond examination behaves like a religious commitment whatever its subject, and calling it one is accurate.

The argument establishes nothing further. Having removed the obligation, it replaces the obligation with a judgment about fit, and it supplies no criterion for making that judgment. Every question about which system a society should have is left open. The argument states this result rather than overlooking it.

Who counts as a countryman

The demands on the form of the polity come in two versions of one program. The versions differ on who belongs.

The newer version defines the nation by its culture, holds that religion is the part of a culture the rest depends on, and concludes that whoever professes the faith is a countryman while whoever rejects the faith has rejected the country.

Rule: countrymen are those who profess the faith.

An Englishman's son loses his faith at seventeen and recovers it at twenty-two.
For five years he is a foreigner in the country he was born in.
A devout Christian in Nigeria is English for all five of those years,
and English again afterward, alongside him.

Two countries adopting the rule hold every member either of them has.

Membership under this rule follows a state of belief that can change inside a single life and can change back. It also awards the same membership to people with no other connection to the place. Applied by several countries at once, the rule stops distinguishing between them, because the criterion is identical everywhere and whoever satisfies it satisfies all of them. The rule then assigns the same population to every nation that adopts it.

The historic version defines the nation as its people, meaning families accumulating into a wider community, and requires that the people, and above all the governing authorities, be Christian. Under that version, a countryman who turns against the faith of his nation does not become a foreigner. He becomes a traitor. The newer version cannot produce that category, because it has no fixed members for anyone to betray.

Separating the two questions works. Membership and allegiance are different questions, and separating them removes the contradiction in the cultural version. That is all the comparison establishes. It does not establish that membership should run by descent, or that holding office should require the faith. Those are further claims, and they are the claims the historic version is actually making.

A state that is several nations

One objection says a nation-is-its-people position cannot function in a multiracial country. The answer given is to redescribe the country. It is an empire ruling several nations rather than a nation itself, which is what the plural in its name refers to. The nations inside it are living groupings rather than administrative ones, so they move, they merge where they are similar enough to merge, and they separate. Nationalism inside such a state is therefore available to any of them, and several have taken it up.

The redescription answers the objection, and it has a cost. A nation of this kind has no border, no membership roll, no institutions, and no procedure by which it decides anything. It can be described. It cannot act. The complaint made about the largest of these nations is that it has forgotten it is a nation and has been outorganized by groups that have not forgotten. That complaint describes the condition rather than a departure from it. A political program addressed to a group with no way of forming an intention has to explain how the intention gets formed. The explanation offered is that the group must remember what it is, which is the thing that needed producing.

The values a legal order carries

Descent alone does not hold a people together. The evidence is a large population that shares descent and cannot act as a unit. What unifies a people is a set of common values, and those values come from a religion. Law rests on those values and culture rests on them. A people split among competing value systems cannot maintain one order, and the argument invokes the principle that a house divided cannot stand. Civilizations containing several religions are not counterexamples. They organized as empires holding several nations under one administration, or they thinned their distinctive doctrines and raised shared secular values until what remained worked as a single religion under a different name. A polity therefore has a religion in fact whether it admits to one or not, and the version that claims neutrality leaves room for value systems that decline the label. On the reading of church-state separation where a nation’s laws must rest on nothing the nation values, the result is not neutrality. The ordering is set by whatever value system fills the space.

The core of this holds, and conceding it costs nothing. A legal order ranks goods, and it has to, because every prohibition prefers one thing over another. No procedure derives that ranking from neutrality, because neutrality is itself a position about which conflicts get left alone. A state that legislates carries a value ordering. Calling the ordering secular describes where it came from and does not show that there is no ordering. This part of the argument stands.

The conclusion claims more than the premise supports. Across the argument, “religion” carries two meanings: a value ordering encoded in a polity’s law, and a confessional faith with doctrines, authorities, and an account of what is true. The premise is strong in the first sense, and the first sense is what makes the historical cases fit, because a civilization that diluted its creeds into shared secular values has a religion only under the wide reading. The conclusion is stated in the second sense: choose the faith and reflect it in the nation’s law. Under the wide reading the necessity claim is true and selects nothing, because every polity already satisfies it, including the ones the argument is aimed at. Under the narrow reading the necessity claim is the thing requiring demonstration, and the case assembled for it was assembled under the wide one.

The step that selects a religion shows the two senses coming apart. Asked which religion, the answer given is the one that is actually true. Nothing before that step was about truth. Unity, the foundation of law, the failure of divided houses, and the fate of civilizations holding several creeds are all functional arguments. Functional necessity licenses at most the creed that will in fact unify the people in question, and for a different people that is a different creed. The argument cannot accept that result, so at the last step the criterion changes to one the functional argument never needed. Either truth selects the creed, in which case the argument about unity was never required to reach the conclusion, or unity selects it, in which case the same argument installs whichever creed a given people already holds.

The consequence is left unstated. The functional argument does not pick a denomination. Applied to a Muslim-majority population it returns an Islamic confessional state by exactly the reasoning that returns a Christian one for a Christian population. Only the truth criterion blocks that result. The truth criterion does the entire selection, and the functional case takes up the entire argument.

The closing either-or carries all of this. Rejecting the historic program is said to be either a rejection of one’s own people or a rejection of Christianity, because both necessities are held to be facts rather than opinions. Those two options cover the field only if the second necessity holds in its narrow sense. Someone can grant that a polity encodes a value ordering, grant that peoples are constituted by descent, and deny that the ordering has to take the form of a confessional establishment. That person has rejected neither of the two things named.

Which passages govern, the early or the later

The peaceable reading of Islam rests on an early verse holding that to kill one man is to kill all humanity. On that reading, modern violence carried out in the faith’s name is a distortion of the verse.

The tradition holds a rule for reading its own text where passages conflict: the later revelation supersedes the earlier. That rule is internal, established in the text and accepted across the sects rather than imposed from outside. Applied to the passages about unbelievers, the peaceable material belongs to the earlier period, and the material commanding war on unbelievers and tribute from subjected People of the Book belongs to the later one. On the tradition’s own ordering rule, the later passages state the operative position, and a polity taking the text as its law takes that position as its law.

A reader outside the faith can check this, because both the ordering rule and the sequence of the passages are internal facts about the tradition. The reach of the rule is not settled. Its scope has been disputed inside the tradition since the classical period, and a reading that confines it to a short list of legal rulings leaves the peaceable material operative. What the argument establishes is that the earlier verse cannot be offered as the faith’s position without an account of why the tradition’s own rule for resolving conflicts does not reach this pair, and that account is owed by whoever offers the verse.

Two supporting arguments do less than the main one. The first is about timing: peace was preached while the movement was small and had no capacity for force, and the position changed once the capacity arrived. That is an observation about circumstance and motive. Taken as an argument, it establishes that a change occurred, which the supersession doctrine already asserts outright. The second is about conduct: the share of religious wars attributed to the tradition, and the representation of its groups in attacks on civilians. That argument offers behavior as evidence of doctrinal content, which the ordering argument had already settled without it. The same form of argument runs in every direction. A tradition judged by the conduct of those acting in its name is judged the same way whichever tradition it is, and the traditions raising the test have records of their own to enter. The doctrinal argument needs the text and the ordering rule. The conduct argument needs an accounting nobody in the exchange proposes to finish.

Conquest as proof of divine backing

Expansion at that speed, against powers older and larger, followed by centuries of intellectual preeminence, is what a divinely backed movement would be expected to look like. Success on that scale requires an explanation, and divine backing supplies one.

The criterion awards the credential to whoever is currently winning, so it keeps reassigning it. The expansion was halted in Europe and reversed over the following centuries. The peninsula where the faith began came under the rule of Christian powers within recent memory. A standard that reads conquest as a mandate has to read those conquests the same way, and the tradition invoking the standard does not accept the result when the direction changes. What survives the reversal is the observation that a rapid conquest requires an explanation, and the ordinary explanations of rapid conquests supply one too.

The treatment of women, measured against what came before

The condition of women in the society the faith arose in was brutal, and the faith improved it: constraints on what men could do, a defined share of inheritance, and limits on practices that had previously had none.

The improvement is granted, and it is not a small concession. The argument stops at the standard it uses. What it establishes is comparative, which is that the arrangement improved on the one it replaced, and a comparative result licenses only a comparative conclusion. Every code that ever replaced another is entitled to the same defense, and the defense returns the same verdict for all of them.

The objection put to that defense is that the comparison belongs against an unchanging moral standard rather than against the paganism in place beforehand. Measured that way, the treatment is severe: discipline administered through violence, plural marriage, concubinage, asymmetric divorce, and standing in law and testimony below a man’s. Read as an argument about political demands rather than about divine origin, the exchange settles little. The comparative defense holds, and it holds only comparatively. The objection runs on a standard the two parties do not share. A reader outside both parties is left with the limit on what the defense covers, and with no verdict.

What the demands have in common

Across both traditions, the demands that survive examination state a constraint and stop there. The ones that fail are the ones that carry a specific institution out of a doctrinal premise.

The verdict on the act of killing follows from the premise. The strategy for reducing the number does not. A rule for treating the foreigner already present is in the text, and the number to be admitted is left undetermined. The criterion for who receives charity survives, while the conclusion that no state can administer it needs a premise about states that is never supplied. Nothing commands a particular form of government, and that argument claims nothing more. The tradition’s own ordering rule fixes which passages govern relations with unbelievers. Reading a conquest as a mandate does not survive the reversal of the conquests. The conditions for a just resistance work as a set of rules, and applying them depends on officeholders willing to be bound by them.

The recurring pattern is a premise that fixes a moral verdict, with an institutional conclusion built on top of it. That conclusion needs a second premise about institutions. The second premise is usually absent, and sometimes nobody notices that it is absent. This is not a defect peculiar to religious argument. It is easier to see in religious argument because the first premise is written down and located in a text, which puts the gap between premise and conclusion where a reader can find it.

How to practice this

1. Take a demand made on the state in the name of a faith and grant its premises in full. Then ask whether the demand follows. Notice that the question can be settled without deciding whether the faith is true.
2. Separate the moral verdict in a demand from the institution the demand asks for. Look for a stated premise about institutions. In most cases no such premise is there.
3. When a demand cites an urgent scale, ask what the scale licenses. Urgency can show the present approach is inadequate. It cannot show which approach is adequate.
4. Check who assesses each condition a demand sets. Notice which conditions only the person proposing the action can assess. Those conditions constrain nobody else.
5. When an argument uses the word religion, check its meaning at each step. One meaning is a value ordering encoded in law, the other a confessional faith. Notice whether the conclusion uses the meaning the premises supported.
6. Put two demands from the same case side by side and check whether they conflict. The greater-evil condition and the argument from scale cannot both be operating.

Related pages

• Validity and Truth: why a demand can be shown not to follow from its premises without touching whether the faith behind it is true.
• Democracy as Sacred Cow: the defenses of democracy examined on their own, which is the work left over once the obligation to it is removed.
• Mass Immigration - Cohesion: the same policy argued from selection and settlement rather than from a scriptural premise.
• LGBT - Permission vs Promotion: the distinction that the argument about the dropped penalty needs and does not supply.

Open questions

All of these arguments are about what a text requires, and none of them is about what a polity did with the requirement. The case this material most lacks is a state-level one: a Muslim-majority polity operating a stated policy of tolerance toward its non-Muslim subjects, with a record of what the policy did under pressure. What would such a case establish that a doctrinal reading cannot, and what would it fail to establish that the doctrinal reading already does?

If a legal order necessarily encodes a value ordering, is there a test that separates an ordering held as a creed from one held as a settlement among people who disagree, or does the difference only appear when the order is under strain?

Sources

What Politics Must Christians Support?, 2026-07-06: https://www.youtube.com/watch?v=1DkdVGxDSso. Supplied the six policy demands and the just-war conditions with the lesser-magistrate route around the authority requirement.

Christian Nationalism Explained, 2026-08-03: https://www.youtube.com/watch?v=CwYzhKZD4Bk. Supplied the two forms of Christian nationalism, the membership case against the cultural form, the empire-of-several-nations redescription, and the argument that a nation requires a religion.

Islam: All Arguments Ranked & Debunked, 2026-04-03: https://www.youtube.com/watch?v=G_9OPDRS5CY. Supplied the supersession argument about the operative passages, the conquest-as-mandate defense, and the comparative defense of the treatment of women.