UPB Discussion

In case you haven’t yet you may want to read my summary of UPB before delving into the discussion below.

A few comments I have on David Gordon’s arguments against UPB:

The heart of the disagreement between Molyneux and me concerns his conception of universality for moral rules. He holds that all such rules must refer to all human beings without reference to particular times or places. He says in his response that he has argued for this requirement, but in fact he hasn’t. He has simply repeated his requirement a number of times.

UPB is the examination of all those theories that propose universal behavioral rules, independent of time and place, nothing more and nothing less. I don’t see why scoping out the subject matter at hand would be something to argue about.

It would be kind of like me saying that I disagree that biology is the examination of theories regarding organic matter. It would be a mere disagreement on preferences about what I think certain words ought to mean, void of any logical or empirical inquiry.

Yet if I were to shoot down every biological theory that the other person proposes on the grounds that it doesn’t also apply to rocks, we will never be able to have a constructive debate. Keep this comparison in mind now as we move forward to examine David’s core objections.

Rape cannot be UPB because sexual penetration is only rape if it is unwanted — thus one man must
want to rape, while the other man must desperately not want to be raped, which means that both of them cannot simultaneously value rape as universally preferable behavior.

Suppose that A is trying to rape B. A wants to rape B but B does not want to be raped. This is entirely consistent with both A’s and B’s thinking that A is morally obligated to rape B. It isn’t a requirement of logic that B want A to do whatever A is morally obligated to do.

First of all, if a theory says person A is obligated to rape person B and thus introduces specific individuals, then it has already stepped outside the bounds of universality, and thus fails to qualify as a UPB rule to begin with. The writer confuses Stefan’s hypothetical test scenario of two people in a room, merely a tool to examine moral theories more closely, with an actual moral (=universal) theory.

Secondly, Stef said in his paragraph that sexual penetration is only rape if it is unwanted by one while wanted by the other party. In other words the term rape by definition requires that one person prefer the act in question and another not prefer it. Thus the theory that rape can be preferred by everybody is in itself a logical contradiction because it uses a word that by definition rules out universal preference. David, in his criticism, switches from “wanted” to “obligated” which changes the intended meaning of the original paragraph and thus leads to confusion.

Thirdly, let’s take aside my above argument and assume the theory in question was a universal theory and the “obligated” vs “wanted” switch didn’t matter: It is still completely immaterial to the validity of a moral theory whether or not A and B both have the ability to believe that it is valid during the act of rape. The entire world could believe that it is a valid moral theory, the truth or falsehood of a theory of universally preferable human behavior depends first and foremost on its internal consistency, not on whether or not everyone could potentially hold it to be valid in their own minds. What Stef was pointing out was a logical contradiction that is implied in the theory in question in and of itself, a contradiction that arises out of the meaning behind the word rape, that is unwanted sexual penetration.

The argument against rape that I found the most philosophically interesting in his book is a different, though related, one. If rape is morally required, then acting against rape is evil, i.e., as Molyneux defines this, morally proscribed. But in order for rape to occur, the victim must resist the rapist. If the victim does not resist, then rape has not taken place. Thus a moral rule that required rape could be put into practice only if evil behavior, i.e., resisting rape, takes place. The purported rule, then, is inconsistent with everyone’s acting as morality dictates.

This argument depends on a questionable premise. If you are obligated to do something, then, plausibly, you can’t also be obligated not to do it. (Some philosophers think that people can sometimes have inconsistent obligations, but I don’t want to appeal to that view here.) But it doesn’t follow from this that everyone else is obligated not to resist what you are doing. Suppose that you are obligated to take care of your child, who needs a medication that only I have on hand. I’m not obligated to give it to you or to sell it to you: perhaps, e.g., someone else also needs the medication. If I do not let you have the medication, I’m not by that fact alone engaging in morally proscribed behavior.

The above confuses the nature of coercive vs. non-coercive behavior. David’s sample theory implies no occupation of anyone’s body against his will (=force). When a theory proposes that a certain behavior is universally preferable and the behavior in question by definition requires that another person not perform that very behavior, or even resist it, then the theory faces a logical contradiction. (I do agree, by the way, with David’s footnote that rape doesn’t necessarily require resistance on the part of the victim since he may acquiesce out of fear or be unconscious/asleep, but I don’t think it poses any problem to the validity the methodology of UPB). Taking care of my child involves no forceful infliction of any behavior upon another passive 3rd party on the receiving end and thus doesn’t qua definition require that another person, say, not take care of his child.

Molyneux says, “This is one of a few arguments against ‘theft as UPB’ in the book — a thief is both violating and affirming property rights when he steals, which is a logical contradiction.”

This argument doesn’t work. A thief is someone who takes what doesn’t belong to him. He wants what he steals, but this does not entail that he thinks he is the legitimate owner of the pilfered goods. To take something is not to make a moral claim to own it. Sometimes people do what they acknowledge that they ought not to do. The thief, one presumes, would not want someone else to take from him what he has stolen; but that does not entail that he would deem someone who did this a violator of his property rights.

Again, this switches the focus from the theory in question itself onto what may or may not go on in the mind of an actor who performs the act that the theory claims universal preference for. What the thief thinks is completely irrelevant! The fact of the matter is that a thief asserts and then retains physical control over an item, combatting potential force in the process and thereafter with force in turn. But the theory of property rights says just that: that it is proper to retain your physical control over things by means of force if necessary.

So if a theory says that theft is universally preferable then it favors the perpetrator of theft by granting him permission to act in opposition to the victim’s property rights, precisely by applying the very concept of property rights in the process. This is the logical contradiction that Stef was referring to above from what I understand.

Since we own our bodies, we also inevitably own the effects of our actions, be they good or bad. If we own the effects of our actions, then clearly we own that which we produce, whether what we produce is a bow, or a book — or a murder.

I am surprised that Molyneux calls attention to a sentence that is evidence of a gross confusion on his part. You cause the effects of your actions and are responsible for them, but it does not follow from this that you “own,” in the moral or legal sense, these effects. What does it mean to “own” a murder?

I’m not sure if David is here asking a genuine question, but in case he is I would like to answer it: Ownership means direct or indirect control over the location of certain matter in space. In the case of a murder this would mean the control over all the relevant tools needed in the process, the coordination of accomplices or the hitman, the slain body, etc.

Again, he may have a different, namely legal, definition in mind but I see no point in getting into a debate about the practicability of terms, unless he thinks that terms were being blatantly misused to manipulate language. In my opinion, from the context, it is clear to me that Stefan is referring to the factual ownership over things, that is the exercise of physical control over matter in space.

The term ownership here is thus of purely observational praxeological nature. Mises has actually done a lot to distinguish the praxeological from the legal meaning of the word “ownership”.

Well, since property rights are a subset of ethics, they must be universal — if universalizing 50% ownership causes ever-declining ownership, clearly the theory has some problems, to say the least. The fact that 50% ownership cannot be rationally sustained is entirely my point. His issue here seems to be with mathematics, not my book.

But, whether rightly or wrongly, I denied that universalizing 50 percent ownership has this consequence. The 50 percent rule is that if I acquire an object, I own half of it. This rule does not prescribe that this half share may be reduced by another half, this quarter share by another half, etc. That is another rule. Molyneux ignores my objection and repeats what he said in his book. One can only echo one of his followers. “Well done, Stef.”

This is a strange argument indeed, and one of us must be missing something very obvious here. I’m saying it’s strange because to me the flaw in David’s line of reasoning seems so crystal clear that I may very well be missing something here.

Stefan said very clearly that UPB examines rules that apply to all people at all times and all places. The conflicting theories that he compares here are:

  • Everyone who obtains a new piece of property should own 100% of that property.
  • Everyone who obtains a new piece of property should own x% of that property, x being anything less than 100.

Using the example of x=50 he explains how this latter rule would create a practical challenge, because every time someone obtains a new piece of property, someone else obtains 50% of that, but then that means that someone else obtains 50% of that particular item, and so on and so forth. This is not necessarily a logical refutation of the theory (it may very well be), but at the very least it is a pretty important practical challenge in for sure!

As far as the origination of ownership, I actually believe that UPB provides a neat framework to establish valid theories in that field. For example, a short version of my approach to that would be:

  • A human being can initially only have gained control over things by means of homesteading land supplied by nature
  • A human being can consciously and willingly surrender control over said property, making it available to other humans
  • A particular version of such a surrender would be a mutual swap of different items amongst two different human beings, also known as exchange
  • All human beings at all times and all places are allowed to retain control over their (homesteaded or exchanged) things by means of force.

The final bold point would be the theory that could be examined and compared to all conflicting theories as to its logical consistency and also empirical prevalence in moral systems and beliefs in general.

You may also be interested in what I wrote in The Ethics of Voluntaryism – An 8 Step “Cross Examination under #5 where I delve into more detail on this.

In any case, I hope that this helps shed some light on what my take on the recent UPB discussion is and of course as always I am open to feedback and correction.

Related Posts:

Universally Preferable Behaviour – A Rational Proof of Secular Ethics

Universally Preferable Behaviour – A Rational Proof of Secular Ethics by Stefan Molyneux from Freedomain Radio is, next to Rothbard’s The Ethics of Liberty, probably the most remarkable book on morality that I have ever read.

Rothbard heavily relied on the conclusions of Natural Law philosophers as a basis to mount his framework upon and derives ethical rules, an “ought”, from man’s nature, an “is”.

Stefan Molyneux completely rejects this approach. Instead he points out that there is in the realm of human behavior really no such thing as absolute, unconditional and universal “oughts”. There is nothing in the pure nature of humans that requires that they be peaceful and good to each other, in the sense that there are indeed physical laws that require that, say, a rock fall down to earth.

One more important thing he points out is this: The main task we need to surmount in terms of establishing a scientific moral framework, is not to evaluate individual actions per se. What we need to evaluate is rather rules regarding actions. The problems of this world are not the petty burglar or killer. Virtually everyone understands the immorality in their acts intuitively. The most dangerous thing are rather ideas about behavioral rules held in people’s minds, general concepts that justify hugely immoral acts under the cloak of morality.

Thus, Universally Preferable Behavior (UPB) is not a framework for evaluating specific actions, but rather one for evaluating behavioral rules.

Understanding Universally Preferable Behavior

Here is how I understood the chain of reasoning. I am mostly taking this from the book, and injecting my own thoughts where I deem appropriate:

1. Reality is composed of objects in the universe, all of which have certain natures, meaning certain specific, and delimitable inputs on them and certain interactions between them yield certain specific, and delimitable outputs (events). These events are, all other things being equal, reproducible or consistent.

2. Logic is the set of objective and consistent rules derived from the consistency of reality:

– Identity: A = A – An object/event is that object/event and not another object/event. A rock on earth is that rock on earth, and not a tree at the same time.

– Non-Contradiction: A AND non-A is false – A proposition that states that something is a thing/event and not that thing/event at the same time is always false. A thing can’t be a tree and not a tree at the same time. An apple can’t fall downward and upward at the same time.

– Excluded middle: A OR non-A is true – A proposition about a thing/event is either true or false. A thing is either an apple or not an apple. An apple either falls down or doesn’t fall down. There is nothing in-between.

3. Validity: A human’s statement about objective reality is a theory. A theory that complies with the 3 laws of logic is valid.

4. Accuracy: A theory that is confirmed by observable evidence in reality is considered accurate.

5. Truth: A theory that is both valid and accurate is true.

6. Preference is the level at which a human being places the desire to perform an action in relation to the desire to perform other actions at any given moment in time. For example at nighttime one prefers sleeping over running. But on the next morning one may prefer running to sleeping. Preferences only exist in people’s minds, meaning they are subjective. Observable human actions, however, are the objective manifestations of subjective preferences. When someone can be observed running then he is showing by his very action that he set out to run because he preferred the act to that of sleeping.

7. Preferable Behavior: When somebody says that some other human being should do something he is making a statement about preferable behavior.

8. Universally Preferable Behavior: When somebody says that all people at all times and at all places should do something, then he is making a statement about universally preferable behavior (UPB), he is proposing a “universal rule”. In short: UPB is any behavior that all humans at all times and at all places should follow. Arguing against the conceptual existence of UPB requires engaging in a debate. But once someone engages in a debate to convince another person, he inevitably implies that all people at all times and at all places should rather prefer truth to falsehood. Once he starts advancing arguments and reasons as to why he is right, then on top of that he affirms that everyone should base his beliefs on universal standards of validity and accuracy. He also affirms that using the same language as your conversation partner is universally preferable. It is impossible to attempt to refute UPB without affirming it in the process. Thus the act of debating and arguing implicitly and inevitably affirms the conceptual existence of UPB.

(I would actually suggest that the commonly known term “Ethics” is a good substitute for the word “UPB”. Molyneux, meanwhile, equates “Ethics” to “Morality”. This is just about semantics, but it does seem to make sense to me and it helps put existing terminologies into context with this new approach.)

9. Morality is defined as the examination of all those universal rules where avoidance of the inflicted effects of the behavior in question would require the use of violence or considerable effort, for example “It is universally preferable to murder.”

10. Aesthetics is defined as the examination of all those universal rules where the inflicted effects of the behavior in question can be avoided without the use of violence and without considerable effort, for example “It is universally preferable to be on time.”

11. The UPB Framework is the process of examining the truth (validity+accuracy) of moral and aesthetic rules. This means that, just as physical or mathematical theories, any true ethical theory needs to be logically consistent (valid) and empirically verifiable (accurate).

Application of Universally Preferable Behavior

Thus there are in general 3 categories that statements about preferable behavior may fall into: morality, aesthetics, or other (all those statements that do not refer to universal, but rather personal preferences). We are here not concerned with those statements that fall in the category other, but mostly interested in morality and to a lesser degree aesthetics, where we have to keep in mind that the differences between the two may not always be black and white, but rather on a fading scale.

Rape: Rape clearly involves the use of violence. Thus any statement about universally preferable behavior involving rape falls into the category of morality. The statement “It is universally preferable to rape.” fails the test of logical consistency. If there are two persons in a room, the statement can’t apply to both people at the same time. One person needs to do the raping, the other needs to be raped. But then the person who is being raped can’t himself rape the other person. Thus the only valid moral statement regarding rape is “It is universally preferable NOT to rape.” or put differently “Rape is immoral.”

Murder: Murder clearly involves the use of violence. Thus any statement about universally preferable behavior involving murder falls into the category of morality. The statement “It is universally preferable to murder.” already fails the test of logical consistency. If there are two persons in a room, the statement can’t apply to both people at the same time. One person needs to do the murdering, the other needs to be murdered. But then the person who is being murdered  can’t himself murder the other person. Thus the only valid moral statement regarding murder is “It is universally preferable NOT to murder.” or put differently “Murder is immoral.”

Theft: Theft involves the use of violence. Thus any statement about universally preferable behavior involving theft falls into the category of morality. The statement “It is universally preferable to steal.” again fails the test of logical consistency. If there are two persons in a room, the statement can’t apply to both people at the same time. One person needs to do the stealing, the other needs to be stolen from . But then the person who is being stolen from can’t himself steal from the other person. Theft also implies the theory that property rights are invalid. But if property ownership rights are invalid it is logically inconsistent to prefer to violently obtain ownership over property, since it is supposedly invalid. Thus the only valid moral statement regarding theft is “It is universally preferable NOT to steal.” or put differently “Theft is immoral.”

In the same manner, many other behavioral theories can be examined using the UPB framework.

Moral Conclusions Regarding Universally Preferable Behavior

The book concludes via this analysis that our political institutions are founded upon inherently and blatantly immoral premises. The idea that “A government is a moral or necessary institution.” by necessity implies that theft is a fundamentally moral action which, as we all know, simply cannot hold.

The military, a group of people sent to another country in green costumes to murder individuals who never attacked them, is of course also an institution founded upon blatantly immoral ideas that are riddled with logical inconsistencies.

The conclusion that I and many other people like Molyneux himself have thus come to is of course that the only moral system is that of voluntaryism.


I believe that that the genius in the UPB framework lies in that it fundamentally and flawlessly explains our natural appreciation for the inherently reciprocal nature of the relation between all elements in the universe, and humans in particular. Logical consistency demands the acknowledgment of this relation. We feel emotionally repulsed against theories about human behavior that fail to recognize this reciprocity, but have been struggling for millennia to explain precisely why that is so.

The UPB framework beautifully integrates the economic concept of value preference into ethics. As far as I know it was the Austrian school’s accomplishment to fully recognize and consistently integrate the notion that value is never an objective or absolute measure, but rather a subjective and ordinal scale where the differentiating operator is simply “better” or “worse”, but nothing like “good” or “bad” or “+/- 100 happiness points”, etc. In that same fashion Molyneux looks at human behavior as nothing but a choice of one action over multiple other actions, and establishes that moral rules are not behavioral absolutes, but rather optional statements about preferable choices, the validity of which, however, remains absolute subject to the laws of logic and proof.

It is my opinion that in this first version, formally and aesthetically Molyneux has unfortunately failed to make this book a pleasant read, in particular for newcomers. The amount of terminological confusions and inconsistencies (I pointed out some here), the abundance of repetitive metaphors, the unnecessary repetition of certain established proofs, and the seeming lack of a consistent and traceable thread at times, really made this relatively short book a tough read for me. To put things into context: This is coming from a guy (me) who enjoyed reading Mises’ Human Action, Socialism, and Theory of Money and Credit with great pleasure! I listened to the audio book twice and read the PDF again before I even remotely felt like I was able to ask qualified and helpful questions.

You will find a lot of criticisms of this book on the net that were written by people who clearly had no real interest in the subject and who deem it necessary to immediately jump on all the terminological weaknesses and inconsistencies that this book is riddled with, rather than being curious and looking beneath the surface. Then there are other criticisms by people who were absolutely and 100% dedicated to understanding the book, but who, in my humble opinion, missed the core aspect of UPB: That it is, just like the scientific method, a scientific framework to examine the validity and accuracy of theories, not of actions, for the simple reason that it is impossible to examine the validity of an individual action.

I think that the actual content, the idea, and the conclusions, when properly understood and connected, are revolutionary, groundbreaking, eye-opening and ingenious. Anybody who is interested in the field of ethics should read this book very carefully and not despair if it doesn’t all get to him as easily as baking pie right away.

Related Posts: