





































Microsoft Word - 2 Oderberg 11.1.docx


[Expositions 11.1 (2017) 15–30]  Expositions (online) ISSN: 1747–5376 
 

Co-operation in the Age of Hobby Lobby: When Sincerity Is Not Enough 

 

DAVID S. ODERBERG 

University of Reading (U.K.) 

 

 

At the time of writing, it has just been announced that the federal Office for Civil Rights has 

upheld the California Department of Managed Health Care’s instruction to health insurers 

covering employees of churches, religiously-affiliated schools and universities, and other 

religious organizations, that they must cover abortion in their plans under the Affordable Care 

Act (ACA)—irrespective of any objections by the plan purchasers.1 

This Californian problem has been rumbling on for several years, and the latest ruling 

reaffirms the state government’s attitude: conscientious objection has no force against the 

provision of abortion via Obamacare. Needless to say, the complaint will run its course through 

the courts, but it seems, given previous judicial rulings, that only the Supreme Court can save 

conscientious objection in California. 

In such a blatant case as this, it is clear that the plan purchasers would be compelled to co-

operate in wrongdoing. (I leave aside the substantive question of the morality of abortion, 

which is another topic. The objectors sincerely believe that abortion is wrong, so let us assume 

it to be wrong for the purpose of analysis.) But what does co-operation in wrongdoing amount 

to, and how to do we judge what kinds of co-operation are themselves morally objectionable? 

The primary distinction is between formal and material co-operation.2 In formal co-operation, 

the co-operator intends the wrong committed by the principal agent, even if he does not intend 

to perform it himself and even if he does not believe the principal act is wrong.3 In other words, 

he shares the guilt of the principal. The Californian case is clear: the plan purchasers do not 

intend either for the insurer to cover abortion or for any abortion to be performed. 

So they count as material co-operators. Here, the relevant principles are those governing the 

doctrine of double effect in general (DDE).4 Why so? Because the agential structure is the same 

as in double-effect cases: the agent intentionally performs an act with both good and bad 

effects, something that may well be permitted if the right conditions apply even though doing 

good and avoiding evil is the fundamental principle of morality. First, what the agent does must 

itself be at least morally permissible. Secondly, the agent must not intend the bad effect itself. 

Third, the bad effect must not itself be a means to the good one, for otherwise the agent would 



Co-operation in the Age of Hobby Lobby  16 
 

 

be using a bad means to a good end. Finally, there must be some kind of proportionality 

between the good and bad effects, sufficient to permit the causation of the latter for the sake 

of5 the former. Proportionality is not essentially a balancing of outcomes as per consequential-

ism, but rather an answer to the question: Are there good enough reasons for permitting the 

bringing about of the bad effect, in terms of what is achieved by production of the good effect? 

Here, it seems fairly clear that the first condition is violated because the act of the plan 

purchaser is to purchase health insurance for employees that covers abortion. As an act, it is no 

more morally permissible than a doctor’s signing a certificate authorizing someone else to 

perform an abortion. But even if the co-operative act (purchasing the insurance) were 

considered permissible, there would be a question over the fourth condition. The bad effect is 

co-operation in the provision of abortion. The good effect is the avoidance of the severe 

penalties of non-compliance with the Affordable Care Act and related state laws and 

regulations. The penalties might be sufficient (let us suppose) to put the plan purchaser out of 

business or stop its effective functioning (a church, say). This is a great evil, but it is not a 

threat to life or limb; abortion, on the other hand, is a direct attack upon life. (Again, we are 

supposing this for the purpose of analysis.) So there seems to be a clear disproportionality 

between the good and bad effects. 

Yet condition 4 is not merely about weighing effects, however this is done. The good effect 

of avoiding the severe penalties for non-compliance is substantial, though not as substantial as 

the evil effect of taking a life. This does not of itself rule out or in the co-operative act. What it 

means is that the co-operator needs a greater reason for assisting the principal than he would 

were the good and bad effects comparable. Environmentalists like to tell us that taking the car 

needs greater justification when the good effect does not significantly exceed the bad effect of 

the extra pollution than when the good effect is substantially greater.6 Compare taking the car 

on a walkable trip to buy milk versus taking your child on a long journey to the hospital. It is, 

of course, easy to imagine cases where the disproportionality is so great that nothing can justify 

co-operation: if I threaten not to be your friend unless you help me beat up an old lady, it is 

hard to see what reason could compensate for the massive lack of balance between effects. 

When it comes to working out what sort of reason can justify certain behavior in a given 

situation, the question in co-operation cases is how the effects relate to the co-operator’s action: 

the more the co-operator is implicated in the principal act, the graver the reason they need for 

co-operating. Someone who edits pornography movies prior to sale, for instance, needs greater 

reason for co-operation than someone who sells the cameras. The former is a more proximate 

co-operator than the latter. Proximity is a factor in implication: it is not only, and sometimes 



17  Oderberg 
 

 

not at all, about spatio-temporal distance between the principal agent and the co-operator. It is 

about the executive character of the co-operative act—in other words whether, morally 

speaking, the co-operator is more like the principal in what they do than someone else actually 

or hypothetically situated in the same chain of events. Proximity is not an absolute matter; one 

is more or less proximate relative to others, or to how oneself or another might have been 

situated. Now entering into a contract to provide insurance coverage for abortion, although not 

as proximate as assisting the abortionist in the clinic, is more proximate than, say, typing up 

the contract and posting it to the insurer. Indeed, in activities where insurable risk is present, 

providing the insurance is one of the key factors that supports the risky activity itself, allowing 

it to be carried out in confidence of protection. One of the motivating ideas behind the ACA is 

that “reproductive rights,” as they are called, be given “seamless coverage.”7 Insurance 

provides the backstop whereby abortion can be carried out without fear of cost or 

administrative burden. Therefore, as well as protecting the act in general, the insurance protects 

particular acts at particular times. This gives the insurance, as with most insurance, an executive 

character putting it very close to the activity itself. 

Moreover, the co-operation looks to be indispensable, in the following way. It is not that 

indispensable co-operation has to be an absolute sine qua non of the principal act, in the sense 

that without it the act simply cannot be performed short of a miracle. It is, rather, that the 

principal act, practically speaking, cannot be performed, or would be performed with great 

difficulty, or would have its probability significantly lowered without the co-operation. 

Providing the burglar with the code to a safe that only you know is as indispensable as can be, 

assuming no other means of access. Handing the burglar the keys to your car is dispensable, 

assuming the burglar could threaten anyone else to do the same. Again, there is a spectrum of 

cases: we are not dealing with mathematics but with ethics, and as we cannot repeat too often, 

Aristotle taught us not to expect more precision that the subject matter allows. 

Now, though, we have a problem: What is it that a hypothetical employee of the plan 

purchaser might want? Is it abortion? Is it abortion without cost-sharing? Is it abortion without 

cost-sharing while remaining an employee of the purchaser? We have to look at all the 

circumstances and ask ourselves what exactly the plan purchaser would be helping the 

employee to do. Two important questions to ask here are: What does the employee want help 

with, and what does the employer think they are helping with? It seems pretty clear that, given 

the purpose of the ACA and of employee health plans generally and given the situation of the 

parties, by purchasing the coverage the employer is helping the employee get trouble-free and 

cost-free abortion coverage as an employee. As such, the employer’s co-operation is well-nigh 



Co-operation in the Age of Hobby Lobby  18 
 

 

indispensable, practically speaking, even though in the current situation the employee could 

make alternative arrangements with difficulty. 

There is a lot more that could be said about the California case and its specifics. My purpose 

has been to show that there is a principled way of looking at co-operation cases that does not 

involve appeal to religious doctrine or expression per se and that is not wholly subjective. Here, 

the employer looks to be a proximate and indispensable material co-operator in wrongdoing by 

providing the coverage, assuming for the sake of argument that the initial act of purchasing the 

coverage is itself permissible (an assumption I in fact deny). As such, it would be morally 

wrong to purchase the coverage, and the state would be compelling the objector to engage in 

immoral behavior. As in U.K. cases, where Catholic adoption agencies have all but ceased 

functioning as such due to government compulsion to pair children with homosexual couples,8 

it would be incumbent on the Californian objectors to submit to the penalties even if it meant 

going out of business or radically changing the nature of their operation. 

The situation is similar to that in Burwell v. Hobby Lobby, where a “closely-held for-profit 

corporation” was granted an exemption from purchasing employee coverage under the 

“contraceptive mandate,” particularly with reference to abortifacient drugs. As is well known, 

the Supreme Court held that the mandate substantially burdened the freedom of religion of the 

employers under the Religious Freedom Restoration Act of 1993 (RFRA). Unfortunately, 

though, matters start to get even thornier than many observers of the current state of play think 

they already are. For whilst I agree with the decision in Hobby Lobby, I do not think the case 

was decided for the right reasons. Yes, the plaintiffs were compelled by the mandate to become 

illicit co-operators in wrongdoing; and perhaps, given the RFRA jurisprudence, the case was 

correctly interpreted. If it was, though, then that jurisprudence is in my view wrong, as I go on 

to explain. In any case, the reasoning in Hobby Lobby has led to the impasse over Zubik v. 

Burwell, in which SCOTUS reversed the previous decisions per curiam and sent the case back 

to the lower courts where, at the time of writing, it lies.9 

The problem centres on the question of how a conscientious objector’s belief about co-

operation is to be understood. Is it a religious/conscientious belief,10 or is it a belief that we 

might call ancillary to a religious belief, perhaps one that is derived from a mixture of religious 

and non-religious beliefs? We need to get clear about the structure of a standard case of Hobby 

Lobby-style conscientious objection. There is a principal act that the objector sincerely believes 

contravenes his/her religious beliefs. Abortion is the most common case post-ACA (including 

abortifacient “contraception”). In other words, taking the Christian context—for it seems to be 

Christians litigating the most frequently—it is an accepted Christian doctrine that abortion is 



19  Oderberg 
 

 

immoral. The objector believes that doctrine. Clearly, then, the belief is a belief that is 

essentially religious in nature, one derived from religious teaching.11 Call this the primary 

belief—the one involving disapproval of the principal act. 

In addition, there is the objector’s belief—call it secondary—that co-operating with the 

primary act is itself immoral. Is this secondary belief itself a religious belief? Well, if the 

objector can show that the belief is somehow embodied in religious teaching, then it too will 

be a religious belief. For example, certain kinds of co-operation in wrongdoing are condemned 

by old-style moral theology manuals of the Catholic Church.12 The condemnation is not 

idiosyncratic, but was (and perhaps is) accepted as standard moral theological teaching across 

the texts used in seminaries, schools, and universities. We can presume that something roughly 

similar applies in many Christian churches; indeed, plaintiffs in the recent cases have been 

either Catholic or of various Protestant denominations. 

Not every kind of co-operation is wrong, however, whether by religious or non-religious 

lights. So it is not as though a plaintiff can simply say, for example: “Co-operation in abortion 

is wrong, as taught by my church. By purchasing employee health coverage for abortion I 

would be co-operating in abortion. Therefore my religious freedom is being violated.” To 

repeat, not every kind of co-operation in wrongdoing is itself wrong. Hence there must be an 

implied tertiary belief involved, e.g., purchasing health insurance for abortion is a wrongful 

form of co-operation. What should we say about this sort of belief? The specific example is 

relatively uncontentious, as I have already claimed: purchasing such coverage (at least as I have 

assumed it to be in the Californian scenario) is proximate, indispensable co-operation and, as 

such, wrong. Now, the theological textbooks agree on the principles, and usually on their 

application, albeit the purchase of health insurance is not exactly a staple of books from the 

1950s and 1960s, where co-operation is discussed extensively. But even if they all said, or if 

some approved religious authority said, that purchasing health coverage for abortion was illicit 

co-operation, in my view this would not itself be a religious belief and protected as such by 

RFRA and the associated jurisprudence. 

Why not? The reason is that the application of religious principles to particular facts is not 

itself a matter of religious belief or teaching, even if all religious authorities make the same 

application in the same way and say so in their texts. More specifically, applying principles of 

co-operation—even ones that are a matter of religious teaching—to the facts of a particular 

case is a matter of logic, prudence, and common sense, not a matter of religious belief. When 

theologians, or religious authorities, apply the principles of co-operation to particular cases, 

they are neither explaining existing religious teaching, nor amplifying it. They are using logic, 



Co-operation in the Age of Hobby Lobby  20 
 

 

prudence, and common sense to derive particular conclusions about particular cases. Now, 

maybe a trained theologian can do this better than the man in the street, but something is not 

ipso facto a theological matter because it is done by a theologian. 

To see this more clearly, we have to recognize that the principles of co-operation, minus 

questions of wrongdoing, apply just as well to cases of perfectly acceptable behavior. For 

instance, I co-operate formally with your mowing the lawn when I plug the brand new mower 

in for you and show you how to use it. You co-operate materially with my buying a six-pack 

of beer by lending me the money because you are my friend, knowing that beer is what I am 

going to buy, but not intending for me to buy it—say due to worries that I might not be sober 

enough for work the next morning. If you have forgotten the password to your computer and I 

alone know it, I co-operate indispensably with you by reminding you what it is. You co-operate 

proximately with my driving to the shop by filling up my car with petrol for me. My co-

operation with your going for a jog is remote when I merely offer to pick up your children from 

school for you. And so on. 

In other words, the application of general principles of co-operation to particular cases 

straddles both wrong and permissible behavior. The application itself is morally neutral, nor is 

it a matter of religious doctrine. One can apply the principles without knowing or caring about 

religion, and without knowing that the principles themselves are theological in nature. In fact, 

although my argument does not hang on this, I doubt that even the principles themselves—e.g., 

that formal co-operation in wrongdoing is itself wrong—are essentially theological. Rather, 

theologians have codified what seems to be part of common-sense moral reasoning, just as the 

Doctrine of Double Effect, despite its codification in religious textbooks, is itself a piece of 

common-sense moral reasoning. 

In any case, recognizing that the application of the principles of co-operation is not itself a 

religious matter is essential to seeing what went wrong with Hobby Lobby and now bedevils 

the theory of co-operation in the wake of that important case. For it is settled law that no court 

can look behind the sincerity of a person’s religious beliefs when the person objects to an 

infringement of their freedom of religion, whether under RFRA or under the Constitution more 

generally. As the majority point out in Hobby Lobby, in respect of the sincere belief of the 

owners of the relevant corporations that providing coverage under the contraceptive mandate 

violated their religious beliefs, “it is not for us to say that their religious beliefs are mistaken 

or insubstantial,” only to determine that the objectors have an “honest conviction” that this is 

so.13 This has long been the legal position as far as religious beliefs are concerned.14 



21  Oderberg 
 

 

If, then, the belief of an objector that by doing X they are co-operating illicitly in some other 

act Y is itself a religious belief, no court is allowed to look behind the sincerity with which it 

is held. In other words, no court is allowed to assess the reasonableness of the belief. Now in 

the case of Hobby Lobby this did not present a problem, which is why the outcome was in my 

view correct. The Greens and the Hahns, owners of the relevant corporations, sincerely 

believed they would be illicitly co-operating with abortion by providing employee coverage 

under the contraceptive mandate. My analysis of the recent California case above would lead 

to the same result: purchasing health insurance to cover employees engaged in wrongful 

activity would not itself be a morally indifferent act. Even if it was, the employers would be 

material but proximate co-operators in serious wrongdoing, without a reason sufficient to 

outweigh the gravity of what they would be involved in. The financial penalties for non-

compliance would be severe, perhaps threatening the survival of the corporations themselves 

and the livelihoods of many people. Yet it is hard to see how proximate co-operation in the 

taking of life can be justified by a pecuniary penalty for refusing. 

What about indispensability? I have assumed in the California case that co-operation is 

indispensable.15 In Hobby Lobby, the Court noted that alternative arrangements for coverage 

could be made for employees using federal exchanges and independent action by health 

insurers or third-party administrators, without employer involvement.16 Against this 

background, the co-operation in Hobby Lobby looks more to be dispensable. Still, the proximity 

of the cooperation and the nature of the evil involved arguably outweigh dispensability. 

Consider the imaginary albeit slightly fanciful case of a large landowner renting his property 

to slave owners. Suppose the government considered slavery to be so important to the economy 

that it required such landowners to provide slavery insurance covering some or all of the tenant 

slave owners’ cost of acquisition and/or maintenance of their slaves. Would we not find the 

purchasing of such insurance to implicate the landowner in so grave a wrong that he would be 

justified in objecting, even if the government were to step in with coverage should the 

landowner refuse to provide it? Moreover, wouldn’t the prospect of the landowner going out 

of business altogether be seen as a price worth paying for resisting complicity in such a wrong? 

Whatever one thinks of the outcome in Hobby Lobby, however, the point is that it was 

decided on the sole ground that the objectors sincerely believed they would be illicit co-

operators. That is, the judgment was that purchasing the insurance “substantially burdened” the 

objectors’ exercise of religion under RFRA, not because they would be illicit co-operators in 

what they sincerely believed, according to religious teaching, to be an immoral act, but because 

they sincerely believed that, according to religious teaching, they would be guilty of such co-



Co-operation in the Age of Hobby Lobby  22 
 

 

operation. Deciding the case on that ground, however, seems to me a grave mistake, whether 

or not it is the right way of interpreting the jurisprudence on religious freedom. 

Why, though, does it matter, given that the outcome was correct? Precisely because of what 

has just happened in Zubik v. Burwell, where SCOTUS vacated the previous judgments and 

sent the parties back to the negotiating table to work out an arrangement that would be judicially 

acceptable. Zubik was the consolidation of a number of lower-court cases, including most 

famously Little Sisters of the Poor v. Burwell,17 in which a religious non-profit organization 

also objected to the contraceptive mandate as substantially burdensome to freedom of religion 

under RFRA. The difference from Hobby Lobby, however, is crucial: whereas in Hobby Lobby 

the Supreme Court held that the objectors (owners of closely-held, for-profit corporations) were 

entitled to the same accommodation granted to religious non-profits, allowing them to opt out 

of providing the relevant insurance coverage, the objectors in Little Sisters of the Poor and then 

in Zubik objected to the very accommodation itself. 

Now, without going into the technical details of how the accommodation—or “opt-out”—

operates, and although it is fair to say that a good chunk of the argument revolved around these 

minutiae, when it is distilled to its essence the objection was against the very idea of opting 

out. In other words, the objectors argued that by having to give notice, either to the Department 

of Health and Human Services or to their insurer or third-party administrator (TPA), of the fact 

that they objected on religious grounds to purchasing the coverage, they were illicit co-

operators in wrongdoing. They claimed that “submitting this notice substantially burdens the 

exercise of their religion, in violation of the Religious Freedom Restoration Act.”18 

Now the Supreme Court declined to decide the case on the merits, no doubt under the 

influence of the recent death of Justice Scalia, leaving SCOTUS at a four-four split in terms of 

liberals versus conservatives, broadly speaking. Be that as it may, and despite the “spin” that 

somehow Zubik was a “victory” for the objectors,19 the fact remains that at the time of writing 

all prior decisions have been vacated and the parties are required by SCOTUS to reach some 

sort of arrangement. Yet the Tenth Circuit Court of Appeals judgment in Little Sisters, which 

effectively no longer stands, is remarkable for the insight it delivers on the case. The Federal 

Court decided by a majority that the opt-out was a “de minimis administrative task” that did 

not substantially burden the objectors’ exercise of religion under RFRA or the Constitution.20 

Unlike the purchasing of coverage, which constitutes compliance with the ACA obligation, 

opting out by giving notice constitutes a way of avoiding compliance with that obligation: it 

“relieves objectors of their coverage responsibility, at which point federal law shifts that 

responsibility to a different actor.”21 As a de minimis task—filling in a form or sending an 



23  Oderberg 
 

 

email—the opt-out was not substantially burdensome either in terms of the complexity of the 

task or, more important, in terms of making the objectors co-operators in what they sincerely 

believed to be wrongdoing. As the Tenth Circuit put it: “Opting out would eliminate their 

complicity with the Mandate and require only routine and minimal administrative paperwork, 

and they are not substantially burdened by the Government’s subsequent efforts to deliver 

contraceptive coverage in their stead.”22 

Yet didn’t the objectors sincerely believe they would “trigger” the coverage simply by 

opting out? Indeed they did, and the District Court in Reaching Souls International v. Sebelius 

emphasised this point,23 which is why it granted an injunction against the government. The 

Tenth Circuit, however, declined to follow this reasoning, but it seems that, given the reasoning 

in Hobby Lobby, they should have! For recall, the objectors’ belief that they would be an illicit 

co-operator was treated by SCOTUS as a sincerely held religious belief that it was not for the 

Court to second-guess. The District Court in Reaching Souls was applying just that reasoning. 

In Little Sisters, however, the Tenth Circuit Court insisted that opting out had the exact opposite 

effect to implicating the objectors: it relieved them of implication in provision of coverage 

under the mandate. The objectors in Little Sisters specifically raised the point that their sincere 

belief they would be implicated by the opt-out had to be accepted as amounting to a substantial 

burden since that belief would be violated by opting out. Yet the Tenth Circuit Court explicitly 

rejected that argument. It was entitled, the Tenth Circuit held, to assess objectively whether a 

law or regulation amounted to a substantial burden,24 which is precisely what it did by judging 

the opt-out to be an eliminator of complicity rather than a facilitator. 

The essence of the judgment comes out well in the way the Court in Little Sisters handled 

the causation question. The objectors argued that opting out of coverage by giving notice, 

whereupon the insurer, TPA, or government stepped in, would “trigger” the coverage itself.25 

As such, the objectors were painting themselves as having a causal role in the provision of 

objectionable coverage under the mandate. Well, if they sincerely believed they had such a 

role, who is the court to disagree, given Hobby Lobby? It looks, on the face of it, that the Circuit 

Court gets itself into knots on precisely this issue. It affirms that “opting out is necessarily a 

but-for cause of someone else” providing the coverage—whether the insurer, the TPA, or the 

government.26 But the Court then goes on to say that the objectors “do not ‘cause’ contraceptive 

coverage by exercising their ability to opt out.”27 So which is it, and why the scare quotes 

around ‘cause’? 

It seems to me the Tenth Circuit Court is wrestling with the problem of giving an objective 

analysis of co-operation in this case while respecting the objectors’ honest beliefs about what 



Co-operation in the Age of Hobby Lobby  24 
 

 

they are doing. The term “but-for cause” is not defined, but presumably the Court means that 

on a simple counterfactual test the objectors, by opting out, do cause the coverage: if they were 

not to opt out, coverage would not be provided by an independent agent, viz., the TPA, the 

insurer, the government, or some combination thereof. But even that is not quite right. If the 

employer were to resist and do nothing, you can be sure that one of the other independent 

agents, i.e., not paid by the employer and not objecting to the coverage, would provide it. If the 

employer were to opt out by giving notice, the same would apply. If the employer were simply 

to go ahead and waive their objection, it is true that no independent agent would step in, since 

it would be the employer themselves footing the bill, with the other agents working with or on 

behalf of the employer to facilitate the process. In the latter scenario, then, the employer would 

be a cause in the counterfactual sense. But comparison with that scenario is not the right one 

to make. By analogy, consider the case of a conscientious objector to serving in the military—

the comparison often made by the courts in these recent cases. Suppose he protested that by 

notifying the government of his objection to serving, he would be a counterfactual cause of 

someone else’s serving, and so would be somehow “triggering” a third party’s participation in 

the military, a form of illicit “but-for” causality. Is this really plausible? What ground could 

the conscientious objector give for this reading of his situation? It seems the only available one 

is that the remaining alternative course of action was for the objector himself to waive his 

objection and be the one to take up arms instead! In other words, if an objector is to be taken 

seriously in their claim of being a counterfactual cause of X, the comparison has to be with an 

alternative scenario in which they are not a cause of X at all, not an alternative in which the 

objector himself would be a direct, personal cause of X. 

Returning to Little Sisters, then, the problem is that the relevant alternative scenarios would 

not prevent coverage being provided, as the Federal Court itself recognized.28 If the objectors 

resist altogether, i.e., do not opt out but sit on their hands instead, a third party will still 

intervene. If the objectors opt in, they will be the cause themselves. So, from the counterfactual 

causation point of view, if there is any co-operation by the objectors it is highly dispensable 

and hence easier to justify than if the opt-out were absolutely necessary for any coverage to be 

provided.29 So what could the Tenth Circuit Court have meant by initially saying that opting-

out was a “but-for” cause of the coverage? Perhaps it had in mind the implicit comparison by 

the objectors to the case where they footed the bill instead. As we have just seen, however, such 

a comparison is of no avail. Or perhaps the Court was merely giving a nod to the objectors’ 

sincere belief that they were a “but-for” cause of the coverage. In the end, however, the Court 

simply could not hold back from giving a dispassionate analysis of the causal situation, nor 



25  Oderberg 
 

 

should it have done otherwise. I think we should interpret the Court as saying something like 

the following: “If the objectors sincerely believe they are in some way causally responsible for 

the coverage should they opt out, they can have their belief. In some bare sense, a ‘but-for’ 

sense involving a comparison on some dimension or other, opting out is a cause. But causation 

is not complicity. You do not become a co-operator, let alone a co-operator in wrongdoing, 

merely by being a bare cause.” 

If that is what the Tenth Circuit meant, they grasped the situation correctly. Opting out is no 

more a form of co-operation than walking away when there is no duty to act, even though by 

walking away someone else will do something to which you object. Of course the devil can be 

in the detail, which was the subject of much analysis both by the Tenth Circuit and in oral 

argument before the Supreme Court. There is no room to discuss all of that here. Suffice it to 

say that in no way was any form of opt-out available to the objectors an authorization, explicit 

or implicit, for anyone else to go ahead and provide coverage. There was no “permission slip,” 

as the objectors complained; on the contrary, as the Tenth Circuit Court observed, “The 

government is not compelling the plaintiffs to endorse or license something they consider 

objectionable; instead, the government is allowing them to decline a legal responsibility while 

requiring another party to perform it in their stead.”30 

The distinction seems to me impeccable. An opt-out, by its very nature, does not meet the 

definition of a co-operative deed, one that assists a principal agent in carrying out a primary 

act. It is certainly not formal co-operation, but neither is it material. The objector cannot say: 

“I am a material co-operator by opting out, since I know that by doing so the principal agent 

will provide that to which I object.” One might as well say that, by running away from a riot in 

order to protect oneself, one is assisting the rioters. Now there might be a good reason not to 

run away, just as there might be a good reason not to opt out: one might be able to help stop 

the riot in the former case, and in the latter one might be able to stop the objectionable primary 

act by not opting out (although, as already noted, this was highly unlikely in the case under 

discussion, except perhaps for TPAs if Justice Baldock’s partial dissent is correct). But that is 

different from saying that by opting out one is a co-operator. Merely being knowingly 

involved, in some bare causal sense, perhaps merely counterfactual, in the occurrence of some 

state of affairs does not make one a co-operator. To be a material co-operator, one has 

knowingly to assist the primary agent; when one opts out, one does not do this. In such a case, 

questions of proximity and the like do not even arise. 

So the Tenth Circuit in Little Sisters accurately assessed the situation and implicitly resiled 

from the reasoning in Hobby Lobby without saying as much. In his partial dissent, however, 



Co-operation in the Age of Hobby Lobby  26 
 

 

Justice Baldock clearly and explicitly recognizes the problem, albeit he does not see it as a 

problem. So we have him citing lower-court judges who held that even if “religious 

organizations are misguided in thinking that this scheme […] makes them complicit in 

facilitating contraception or abortion,”31 that is not for the courts to second-guess. Justice 

Baldock adds: “Hobby Lobby supports this position well.”32 Nevertheless, he does not pursue 

this issue and goes on instead to focus on a possible technical distinction between insured 

plaintiffs, where he agrees with the majority, and self-insured plaintiffs using TPAs, where he 

argues the plaintiffs’ sincere beliefs align with reality. 

For my purposes, the point is that some judges do think that mere sincerity is enough when 

it comes to RFRA cases, and as a matter of fact Hobby Lobby supports this view. I submit that 

this is an absurd position to be in. One might salvage the situation by arguing that, just as Justice 

Baldock did with sincerity and reality in the case of self-insured plaintiffs, so in Hobby Lobby 

what was really going on was that purchasing coverage, to which the owners objected, was 

clearly not morally neutral, as I suggested earlier in discussing the situation in California. On 

this line of reasoning, the sincere belief of the owners that they would be illicit (material) co-

operators was grounded in reality, and so no deep analysis was needed of what the conditions 

for illicit co-operation actually were. 

Yet this is mere speculation. On the face of it, Hobby Lobby does support the “mere 

sincerity” view, which is troubling because it strictly does allow that RFRA applies to a sincere 

belief that an opt-out can be a means of illicit co-operation.33 If sincerity is enough, then by the 

reasoning in Hobby Lobby a court might allow a RFRA claim by an employee of a non-

objecting corporation, who was required on pain of dismissal to take the company mail to the 

post office—including health insurance contracts the employee found objectionable. Although 

the employee’s co-operation would be both remote and dispensable, her sincere belief that she 

was an illicit co-operator could not be gainsaid on the most obvious interpretation of Hobby 

Lobby. 

The courts should not allow sincerity to be enough. What they have to do, and what the 

Supreme Court, for reasons one can understand, ducked in Hobby Lobby, is to articulate some 

general, reasonable, and objective principles of co-operation that would enable a dispassionate 

analysis of where a conscientious objector might stand, whatever their sincerely held beliefs. 

Justice Alito was quite right to assert in Hobby Lobby that the ethics of co-operation represent 

“a difficult and important question of religion and moral philosophy, namely, the circumstances 

under which it is wrong for a person to perform an act that is innocent in itself but that has the 

effect of enabling or facilitating the commission of an immoral act by another.”34 Difficulties 



27  Oderberg 
 

 

notwithstanding, courts need to take notice of what both philosophers and theologians have 

said about this very question. Taking notice of moral theology does not entail assessment of 

the merits of a distinctively religious belief, nor does it entail adopting any particular religious 

position. Rather, it means recognizing the philosophical merits of the hard work put in by some 

very clever moral theologians over the decades, and using the fruits of that labor to bring some 

order into what currently threatens to be judicial chaos. With freedom of religion and of 

conscience on the line as never before, the least the courts can do is to give objectors a clear 

pathway through the legal thickets. Those asserting their RFRA rights and who are not, 

objectively, illicit co-operators in wrongdoing still have many other avenues to make their 

protests heard, such as the time-honored ones of pressure on Congress, media publicity, and 

civil disobedience to unjust laws. 

 

 

Notes 

 

1. See http://www.adfmedia.org/files/CDMHCInvestigationClosureLetter.pdf. 

2. My original paper on the topic is “The Ethics of Co-operation in Wrongdoing,” in 

Modern Moral Philosophy, ed. Anthony O’Hear (Cambridge: Cambridge University 

Press, 2004), 203–27. This paper was cited in the majority judgment of the U.S. 

Supreme Court in Burwell v. Hobby Lobby (573 U. S. [2014], slip op. at 36, n. 34). 

3. Hence a formal co-operator does approve the wrongful act, but can do so even if they 

don’t see it as wrong. A person might think it just fine to rape someone as an act of 

revenge, say, and so approve the act, without thinking the act wrongful. In other words, 

approval of the wrongful act does not entail thinking of it as wrong, and in fact usually 

occurs in conjunction with the belief that it is somehow justified. 

4. For an outline and defence of PDE, see my Moral Theory (Oxford: Blackwell, 2000), 

ch. 3, as well as my “The Doctrine of Double Effect,” in A Companion to the Philosophy 

of Action, ed. T. O'Connor and C. Sandis (Oxford: Wiley-Blackwell, 2010): 324–330. 

5. To repeat, not as a means to. 

6. A classic double-effect case. One can think of similar cases if this sort of example does 

not convince. 



Co-operation in the Age of Hobby Lobby  28 
 

 

7. The term “seamless coverage” is used in connection with the “contraceptive mandate,” 

but obviously applies equally to abortion, as far as the government is concerned (and 

note that objectors to the contraceptive mandate have sometimes focused on the 

abortifacient contraceptives in particular, as in Hobby Lobby). See oral argument in 

Zubik v. Burwell, http://www.supremecourt.gov/oral_arguments/argument_transcripts/ 

14-1418_1bn2.pdf, at pp. 19, 20, 32, 42, 48, 60, 83. Accessed July 4, 2016. 

8. Some have closed while others have disaffiliated from the church in order to continue 

providing gay adoption and hence maintain charitable status under the law. See 

http://www.catholicherald.co.uk/commentandblogs/2013/07/04/how-many-catholic-

adoption-societies-have-actually-closed-down-and-how-many-are-now-quietly-

handing-children-over-to-gay-adoptive-parents/. Accessed July 4, 2016. 

9. See http://www.supremecourt.gov/opinions/15pdf/14-1418_8758.pdf. Accessed July 

4, 2016. 

10. I will usually refer only to “religious” belief since the bulk of the jurisprudence in the 

area concerns religious freedom, and after all the key legislation is the Religious 

Freedom Restoration Act. Nevertheless, the same issues apply to conscientious 

objection in general, where the grounds are generically ethical rather than specifically 

religious. I will, however, continue to speak of “conscientious objection” as well as 

“religious objection” since the former is standard terminology even in the religious 

context. 

11. Whether or not the objector also believes abortion is wrong for non-religious reasons. 

12. An excellent overview of the teaching can be found in J.A. McHugh and C.J. Callan, 

Moral Theology: A Complete Course, vol. 1 (New York: Joseph F. Wagner, Inc., 1958), 

615–641. The references by Justice Alito in Hobby Lobby (slip op. at 36, n. 34) are also 

pertinent sources. 

13. Hobby Lobby, slip op. at 37–38. 

14. Thomas v. Review Board of the Indiana Employment Security Division (450 U.S. 707 

[1981]); Hernandez v. Commissioner of Internal Revenue (490 U.S. 680 [1989]). 



29  Oderberg 
 

 

15. Whether it really is does not matter for present purposes; in any case, the availability 

of coverage without employer co-operation is hard to work out given the changing 

circumstances in which the ACA is being implemented. 

16. Hobby Lobby, slip Op. at 9–10, and n.8. 

17. See http://www.becketfund.org/wp-content/uploads/2015/07/LSP-Op.pdf (2015). 

Accessed July 6, 2016. 

18. Zubik, slip op. at 3. 

19. See, e.g., http://www.heritage.org/research/reports/2016/05/supreme-court-decision-a-

huge-victory-for-little-sisters. 

20. Little Sisters, slip op. at 42, 45, 50, 51, 78, 79, 84 (header pagination). 

21. Ibid., 60. 

22. Ibid., 82. 

23. 2013 WL 6804259, at http://www.clearinghouse.net/chDocs/public/FA-OK-0002-

0001.pdf, pp. 13–14, quoted in Little Sisters, slip op. at 36. 

24. Little Sisters, slip op. at 48, n. 23. 

25. Ibid., 55. 

26. Ibid., 66. 

27. Ibid., 68. Both quotations occur specifically in the context of self-insured employers 

and their TPAs, since Justice Baldock’s partial dissent (pages 109–133), whilst 

agreeing with most of what the majority held, distinguished the case of TPAs. I will not 

get into the details of that issue; the point is that the Court is quite clear that what it says 

in these quotations applies a fortiori to employers who are not self-insured via TPAs. 

28. Ibid., 68. 

29. As Justice Baldock, partially dissenting, thought was the case with self-insured 

plaintiffs and their TPAs. 

30. Little Sisters, slip op. at 66, n. 36. 



Co-operation in the Age of Hobby Lobby  30 
 

 

31. Justice Baldock, in Little Sisters, slip op. at 111, citing dissent of Justice Kavanaugh in 

Priests For Life v. HHS, No. 13–5368 (2015), slip op. at 35, available at 

http://cases.justia.com/federal/appellate-courts/cadc/13-5368/13-5368-2015-05-

20.pdf?ts=1432159255. Accessed July 7, 2016. 

32. Little Sisters, slip op. at 111. 

33. Assuming that it is a genuine opt-out, i.e. that it is not something else in disguise. And 

we must add the usual “all things being equal” rider. 

34. Hobby Lobby, slip op. at 36, n. 34. 


