The Retributive Theory of Rights

There are many sorts of rights to many different kinds of things and most rights are complex combinations of other rights.  The standard taxonomy of basic rights comes to us from Wesley Hohfeld.  Hohfeld argued that all rights are, or are combinations of, what he called claims, duties, privileges, powers and immunities.  We’ll call them Hohfeld rights (RightsH).

Hohfeld RightsH

A has a claimH that B φ


B has a dutyH to A to φ

A has a privilegeH, with respect to B, to φ


A has no dutyH  to B not to φ

A has the  privilegeH of  φ-ing 


A has no dutyH  to anyone not to φ.

A has a powerH


A has the ability to alter her own or another’s RightsH.

B has an immunity


A lacks the ability to alter B‘s RightsH

Retributive rights (RightsR) are Hohfeld rights to do or not to do a particular kind of thing; viz. to forcibly interfere with other people to prevent them from acting, or to make them act, in certain ways.  Self Defense is a paradigm example of a Retributive right.  Someone acting in self-defense is free, with respect to his attacker, of the normal duty not to harm another.  He has  what Hohfeld would call the privilege of fighting back. 

We can define all the forms of rightR in terms of the Hohfeld’s privilegeH:

Retributive RightsR

A has a claimR,  against B, that B φ

B has a dutyR to A to φ


A has a privilegeH, with respect to B,  to force B to φ

A has a claimR,  against B, that B not φ

B has a dutyR to A not to φ


A has a privilegeH, with respect to B,  to force B not to φ.

A has a privilegeR, with respect to B, to φ


B has no claimR, against A, that A not φ

A has no duty
R  to B not to φ

B does not have a privilegeH with respect to A to force A not to φ.

A has the  privilegeR of  φ-ing 


A has no dutyR  to anyone not to φ

No one has a claim
R against A’s φ-ing

No one has the privilegeH to forcibly prevent A’s φ-ing

A has a powerR


A has the ability to alter her own or another’s rightsR

B has an immunityR


A lacks the ability to alter B‘s rightsR.

Here we understand  “forcing  B to φ ” as  matter of applying the minimum violence or harm  required to cause B to φ.  If you use more force than is minimally required you are doing more than just forcing B to φ; more than you have a right to.  In the case where B will φ without any coercion by A then the minimal force is, of course, no force at all.  

The right of self-defense now appears as the claimR that others not harm you, since.

A has a claimR that B not attack A

B has a dutyR not to attack A


A has the privilegeH, with respect to B, to forcibly prevent B from attacking  A.

In my last post  on Retributivism I tried to demonstrate that the rights conferred by promising are Retributive.  When A promises B to φ, I argued,  what  is going on is that A is exercising his powerR to waive his normal claimsR  not to be interfered with to confer on B a claimR that A φ.

Hohfeld himself spoke of another Retributive right which will be of special interest to us.  He speaks of a right to “non-interference” in performing an action.  Such a right might arise, say, if B promises not to interfere with A’s φ-ing.  

B’s promise gives A a claimH against B that B not interfere in with A’s φ-ing.   Given that:

A has a claimH against B that B not interfere with A’s φ-ing


~(B  has a privilegeH, with respect to A,  to force A not to   φ.)


~(B  has a privilegeH, with respect to A,  to force A not to φ.)


~ B has a claimR that A not φ

and, in turn,

~B has a claimR that A not  φ


A has the privilegeR, with respect to B, to  φ


B has a dutyH to A not to interfere with A’s φ-ing


A has a privilegeR, with respect to B,  to φ.

So, Hohfeld’s right to non-interference in one’s performance of φ  comes down to a privlegeR to φ.

Retributive Ethics

No one, at least no one willing to acknowledge moral rights at all, is likely to deny the existence of Retributive rights. But Retributive Ethics is not just the doctrine that there are rightsR.  It is the thesis that all rights are Retributive:  that they all are, or are combinations of, claimsR, dutiesR and privilegesR.

This is a substantive claim. One, in effect, denied by Hohfeld and more recently by Judith Jarvis Thomson.  They believe that there are rightsH  that are not rightsR.  Let’s look at their arguments.

In a notoriously perplexing passage, Hohfeld invites us to imagine that C has purchased a shrimp salad and D is hungry. Hohfeld says,

“C being the owner of the salad, might say to D: “Eat the salad, if you can: you have our license to do so but I don’t agree not to interfere with you.” In such a case the privilege exists, so that if D succeeds in eating the salad, he has violated no right of C’s.  But it equally clear that if C had succeeded in holding so fast to the dish that D couldn’t eat the contents, no right of D would have been violated.”*

C has said, in other words,  “You can eat my Shrimp salad when you pry it from my cold dead hand.”  C is giving D permission (a privilegeH )– to eat the salad but  C is not promising — and hence not giving D a claimH that C will not interfere with D ‘s attempt to eat the salad.   It is supposed to be a case where, to put it in our terms.

D has the privilegeH  with respect to C that D φ  but D does not have privilegeR with respect to C that D φ.

Such a possibility is certainly on the cards given the austere definitions of the terms.  To have a privilegeR  to eat the salad to is to have a claimH that others not interfere with your eating, it does not entail that there are no claimsH against your eating.  Conversely, C might waive his claimsH against D’s eating the salad—thus giving D the privilegeH of eating it— without waving his own privilegesH to defend his salad to the death.  

But does this logical possibility correspond to any plausible moral reality? .  The salad is in my house and I give you the privilegeH of eating it (=I waive my claimH that you not eat it),  but I won’t disable the burglar alarm,  tether the Doberman or lower the drawbridge and I have, by the way, told my sentries to shoot on sight. What “right” have I given you exactly?

Thomson concedes that this sounds odd.  She agrees that normally when we give someone permission to, say,  help themselves to our shrimp salad we are, whatever else we are doing,  promising not to interfere with their doing so.  Normally, she says, when we give permission we are giving what she calls a “liberty” which she defines as the combination of a Hohfeldean privilegeH and claim to non-interference.  It is, what she calls, a “cluster right”.  In our terms

D has the  libertyT  to φ  =  D has the privilegeH to φ  and the privilegeR to φ.

But this formula only formalizes the riddle.  If you have the PrivilegeR of doing something– if no one has a right to interfere with your doing it.  Why do you also need a privilegeH?  Note that when C gives D the right to non-interference — the PrivilegeR— of eating the salad, C is not just saying that he won’t interfere, he is renouncing all right to do so.   It as if I  say to you, “This is my salad.  You have a duty not to eat it without my permission. And I do not give you permission to eat it.  However I hereby waive any right to interfere with your eating that salad.  Indeed and if  I do anything that in the least impedes you on your way to the salad I herewith promise to plead guilty if you charge me with assault.”  What now stands between you and the salad?  What have I withheld?  Conversely,  why would anyone want a PrivilegeH absent the corresponding PrivilegeR.


Thomson has an answer.

“It is not easy to imagine a background against which it can easily be understood why a man might be interested in getting the privilege, if he knows he is going to get only the privilege.  What use is it to D to get the privilege of eating the salad, if C may well decide to hold “so fast to the dish that D can’t eat the contents” ? — and, moreover, to do this without in any way wronging D? Presumably the use to D is this: if C does not interfere with D’s eating of the salad, and nothing else does either, so that D is able to, and does, eat the salad,  then C cannot complain that D wronged him.  (As Hohfeld put it: “he has violated no right of C’s”.)  That is certainly not nothing. If you have not given me the privilege of eating your salad, and I manage to eat it anyway, then you can complain that I wronged you.”  *

Thomson is on to something but her way of putting things is very odd.  Notice in the first place, that, taken literally, it is just false.  Unless C gets his jaw broken in the scuffle over the Salad,  C can complain about D’s eating his salad whatever permissions C gave or didn’t give.   Believe it or not, there are people in this world who will offer you their salad but then, when you take them up on it,  will whinge all day long about what a nice salad it was and why did you take them up on their offer and  how much they wish they had their bloody salad, blah, blah, blah…

People can  complain about anything.  But note that Thomson here, on pain of circularity or worse, must avoid the more natural formulation “has a right to complain” to avoid the awkward question of what sort of right that might be and where it might come from. 

Then too there is the puzzle about what use is it to complain? Thomson says that it is “certainly not nothing”.  But, if not, it comes perilously close to being nothing.  Complain to whom? The guy who ate your salad? But for what purpose? To annoy him? For revenge?  And suppose that you do complain to him, or to the courts or the King or God himself, and they listen to you with maximum indulgence, what good is that? What is the point of your crier de coeur?  What do you expect them to do?

There is  one obvious answer:  you want compensation.

Remember it was part of the original story that the salad was D’s property.  Property rights are cluster rights.  They include not just the right to defend your property but rights to compensation from anyone who deprives you of your property, whether or not you defended it.  Normally giving someone your salad is not just promising not to interfere with them,  it is also waving your rights to compensation.   But in this story,  C has only waived his rights to defend the salad, he has not waived his claim to compensation. So he can “complain” the sense of “bringing a complaint” in law. He can press a justified claimH before the courts. 

 So D does have an extra claimH he can waive– and additional privilegeH he can give– beyond the PrivilegeR  to eat the salad.  But if it’s a right to compensation Thomson has in mind  it doesn’t sustain her argument.  Remember that she was trying to show us that a privilegeH to do φ   was a morally significant “something”.  Something  separate from and additional to the privilegeR to do that thing.  If so, then  privilegeH to eat the salad must preclude some other sort of claim, against D’s eating the salad.  But C’s claims to compensation do not fit that bill.  The privilege that D would get if C waived his right to compensation is not a privilege to eat the salad it is a privilege not to compensate D for eating it.

So here is the state of play. Because the salad is C’s property,  D has (at least)  three kinds of duties.  D has the dutyH not to take the salad (unless C waives that duty) and D has the dutyH to compensate C if he does eat the salad (unless that is waived) whether or not C tried to interfere.  Additionally, because it is his property, D has a dutyR to C not to eat, that is to say, C has the right (the privilegeH)  to defend his property against C .

(1)    D has a dutyH to C not to eat the salad.

(2)     D has a dutyH to C to compensate C if D eats the salad.

(3)     D has a dutyR to C not to eat the salad  ( = C has the privilegeH  with respect to D, to interfere with D’s attempt to eat the salad.)

If C waives (1) then D gets privilege (1’).  If C waives (2) then D gets privilege (2’).  If C waives D then he gets privilege (3).

(1’)    D has a privilegeH with respect to C to eat the salad.

(2′)    D has a privilegeH with respect to C not to pay compensation.

(3′)    D has a privilegeR  with respect to C to eat the salad.

Thomson is trying to show that C has more rights vis a vis the salad, than are captured by (3).   She is correct that there is more to C’s property right than (3) insofar as it includes, at least, the claim to compensation (2) against anyone who eats it.  But (2) is not (1).  In the first place they are duties to do different things.  In the second place it seems you can give someone permission, in the sense of a libertyT, to eat your salad without waving their duty (2).  This seems to be what happens when you say something like.  “You can touch it, but if you break it, you’ve bought it.”. 

Thomson’s argument was that if he does not waive (1) then C has grounds for complaint if D eats the salad because he hadn’t given D  privilege  (1).  This was supposed to show that the privilege  (1) was morally significant, because C still had grounds for complaint.  But now we can see there is another right in play.  It now seems that even if C had waived (1) and (3), C would still have grounds for complaint (until he is compensated) because C hadn’t waived (2) and  because  D has failed in his corresponding  dutyH to compensate C.  

So let us remove (2) from equation.  Imagine that C waives both (2) and (3), by waiving his right not to interfere with D’s attempts to eat the salad and waiving all claims for compensation from D in the event that D does eat the salad.  Let that exclude even compensation for the pain of remorse filled second thoughts.  That should still leave D with a DutyH not to eat the salad.  But what is now left?  If D now eats the salad, C’s right (1) has been abridged, and I suppose he can complain, but the complaint now does seem pointless whining.  C can’t expect any compensation from D, he is not owed any.  He gave up that right when he waived (2).  The salad is gone forever.  It’s too late to forcibly defend it, so (3) is beside the point.  He might as well complain about the weather.  C’s claimH that D not eat his salad,  now seems morally empty.

Or so it seems to me.  Am I saying it makes no difference whether C gives permission to D to eat his salad?  No!  I think it matters very much.  But it matters because, in real life, what we are doing when we give someone else permission to use or consume our property we are doing at least two things. 

First Thing: We are waiving our rights to forcibly defend our property against the other’s using or consuming it.  We relieving people of their dutiesR to us to leave our property alone.

Second Thing: We are waiving our rights to compensation if the other uses or consumes the property.  Of course,  for the Retributivist, the right to compensation will be a rightR:  It will be the right to force the other to perform some compensating act.

If Thomson were right there would have to be a third thing.  There would have to be a separate rightH – the one expressed in (1) — that would have to be waived before we were giving permission in full.  But, we have not seen any evidence for it.  It’s not clear what’s left when one has done the first and the second thing.

It is possible for someone to protest that the Retributivist has got it wrong about the rights to compensation, and that it wasn’t on the cards to waive (2) without waving (1).   This it turns out is something like, Thomson’s position.  She thinks that  when D fails in his duties like (1)  to C this can give rise to  duties like (2).   Breaching a right, she thinks,  always leaves a  moral trace or “residue” that can include a duty to compensation. 

Before we can consider this we’ll have to switch examples.  The problem with  Hohfeld’s salad case is that involved the idea of property.  Property rights, on anyone’s account, are compound and complex, and likely to confuse.  Especially complicating is the fact that our rights to compensation for the destruction of our property are different from our more general rights to compensation.  We have a general right to compensation when someone harms us.  But we have it seems rights to compensation for the destruction of our property which are independent of any harm done to us. 

Thus.  Here is a tree.  I chop it down and, because I carelessly forget to yell “Timber!” you don’t get out of the way and it falls on you. I owe you some sort of compensation for the injuries I have caused you.

Now tell the story differently. I am more careful this time.  I wait till you are far away.  I chop down the tree.  You are physically unharmed.  Moreover, I know you have always hated that tree and will be delighted that it is gone.  Apparently, I don’t owe you compensation. 

But wait! Here’s another datum.  It was your tree.  It was your property.  Now, once again, I owe you compensation, and it really doesn’t matter if you are not harmed at all in any non-proprietary sense.  Suppose  it’s true that you always hated that tree for obstructing your view and that you had no use for the lumber, and would happily have paid me to chop down the tree and haul it away if I had offered.  You are delighted that it is gone!  Still, if I didn’t get permission it might seem that, morally speaking — and certainly legally–  I owe you compensation—fair market value say– if you are churlish enough to insist on it.

Property rights are peculiar and interesting.  In a future post I’ll begin a Retributivist account of them.  But for now, to sort out the issue between Thomson and me, we need a simpler kind of case.

The rights conferred by promising will serve our purposes.  Thomson thinks, and I agree, that promising does confer rights: if D promises C to do φ then D has a duty, to C, to φ

So let this be our story. The salad belongs to no one.  It is free for the taking to anyone who comes along.  D comes along, he is hungry, but for some reason, D has promised C that he would not take the salad.

D goes ahead and takes the salad anyway.  What follows?

In the realm of rights Thomson uses cases like this to argue that rights are not “absolute” in the sense that it is not always morally wrong to violate them.  Thus, we can imagine that D has very good reasons—morally persuasive reasons–  for eating the salad.  Suppose that since D made the promise he has fallen on hard times and now will starve to death if he does not eat the salad.   Or  suppose that D needs the salad to feed his starving children when he knows that C is childless and well fed.  In such a case, Thomson says, it would not be morally wrong for D to take the salad but, she  adds, that does not mean that D’s duty to C goes away.  There is, a moral residue or trace, which remains from the promise breaking  and it takes the form of a new right: a claim that D compensate C for taking the salad, however morally justified that taking may have been.

If this leaving of traces or residues is a real feature of rights then it is not something that the Retributivist can account for.  To mix in other metaphors, on Thomson’s view a duty is like an egg, when you break it a new duty springs forth.  Breaching a promise, begets a duty to compensate.

Retributive rights are not obviously protean in this way.  On the Retributivist account of promising, when D makes his promise it becomes true that:

D has a dutyR, to C, not to take the salad


C has a claimR that D not to take the salad.

Both of these amount to saying that D’s rights would not be violated if C forcibly prevents him from getting the salad.  On the Retributive view  when D promises C  not to φ, what D is doing is giving C permission to forcibly prevent him, if necessary, from φ-ing. 

But once the φ-ing is over.  Once D has eaten the salad, it doesn’t follow from the fact that D has breached his DutyR that he now has any sort of duty to compensate C.  

All of which would suggest that there is more to duties, properly so called, than dutiesR.

But is it so?    Listen to Thomson’s explanation of how the breach of a promise leads to a right to compensation.

“… there remains a central similarity between informal promising and formal contracting.  Making a promise, like signing a contract, is positively inviting reliance. 

And, where I invite your reliance on an expectation about my future behavior.  I thereby take on myself a complex responsibility, namely to make that expectation true or, if I will for some reason be unable to, to take reasonable steps to see that you do not lose by virtue of having accepted my invitation to rely.”*

But now does C’s right to compensation in any given case arise from the fact that the promise has been broken, or does it arise from the fact that by his act of promising,  D may have lead C  to rely on an expectation of C’s behavior?  As Thomson herself notes, promising is not the only way of leading someone to rely on their expectations about what you will do. 

Suppose that every day as long as either can remember there has been a shrimp salad for the taking.  But D has never taken it and indeed,  D has told C many times that he is deathly allergic to shellfish of any kind.  So C comes to rely on D’s not touching the salad, and D knows that C relies on  him to behave this way and why.   Nevertheless, one day, without warning, on a whim,  D decides to take the salad  to offer as present to some other, non-allergic friend.  C must go without his lunch .   Now it seems C has grounds for complaint.  D had knowingly lead C to rely on him to behave in a certain way.  D owes C compensation—he should at least buy C lunch.  C has the same rights to compensation he would have if D had explicitly promised never to touch the salad.  But D didn’t make such a  promise and arguably D had no duty not to take the salad, or if he did, it stemmed not from any act of  promise giving but from a prior obligation not do anything to harm another.

Now tell the story another way.  Suppose that D does explicitly promise never to touch the salad.  But now D breaks his promise, for reasons entirely beyond his control. Or, as Thomson says:

“Or of course the victim of the breach of promise may suffer no loss at all, or only a loss not worth troubling about to make compensation for, perhaps because he or she did not rely on the promise, perhaps because the cost of the frustrated reliance was so small.”*

In such cases, Thomson thinks, no compensation is owed.  Of course she’s right, but notice that in these cases it is no less true that a promise has been broken and the promise breaker has failed to do what he had a duty to do.   This shows that  the right of compensation does not spring from the breach of promise per se.  That it is not true that.

(T)       If D has a duty to C to φ and D does not Φ then D has a duty to compensate C for not φ-ing.

Duties to compensate arise, when they do,  when we harm others, whether or not we have antecedent duties to them.  They spring, as it were, from broken heads not broken vows. 

Remember we started down this path in pursuit of the moral weight of non-retributive claims and duties.  Once I have  promised not to interfere with your φ -ing—once I have given you the PrivilegeR to φ —  what more  do you need or want?  If you have no dutyR to me not φ.  What  do you have that can constitute a dutyH

It cannot be a dutyH to compensate me if you φ, since, that is not a dutyH to refrain from φ -ing, but to do something else (compensate me  if you φ).   Nor is it the thing which produces the duty to compensate me if you φ.  As we just saw, even Thompson thinks that the duty to compensate arises only when (willful) harm is done and it seems you always have duties to compensate someone when you willfully harm them by φ-ing  independently of any duty the promise may create.

The  universal negative cannot be proved, but, for all we have seen, there is no evidence that there are is anything more to rights that rightsR.

Retributive rights are Hohfeld rights in the sense that they obey the logic of Hohfeld’s claims and duties and privileges.  But Hohfeld’s rights are empty schema.  To have a Hohfeld privilegeH to φ is just for there to be no one to whom one has a duty not to φ.   As Thomson points out,  this entails that, an inanimate object,  because it has no duties at all, has the privilege to do anything at all.  By this measure, your shoes have more rightsH than you.  

Retributivists rights have morally consequential contents.  The retributive privilegeR to φ does entail a claim,  not to φ, but rather not to be forcibly interfered with in φ.  Retributive rights  can  be always be cashed  out in the coin of  the kinds of interpersonal violence they sanction or enjoin.  Retributivism– the view that all rights are retributive–  requires that we do this accounting for all our moral talk.

And when we run over our libraries, persuaded of its principles, what havoc must we make?  If we take in our hand any volume of ethical theory or political philosophy let us ask:  Does it assert principles about who may be forced by whom, to do what and when?   If the answer is no we should consign it to the flames for, it can then contain nothing but moralistic sophistry and illusion.

Leave a Reply

Your email address will not be published. Required fields are marked *