First separation of 'is' from 'ought'   
The logical separation of 'is' from 'ought', of facts from values, of causal relations from normative connections, is the most important modern criterium for science.
The first requirement of this criterium is that norms and values are not deduced from facts. Kelsen tries to meet this requirement by developing an argument that can be seen as one example of the familiar way in which moral objectivism is criticized. Therefore, I will discuss this argument rather briefly .
Very often Kelsen simply writes as if the modern scientific worldview is true, that is, as if we know for certain that reality contains nothing but facts and regular conjunctions of facts. But Kelsen also admits that this worldview is not shared by those who believe in objective, naturally given, supreme values; that is, by those who believe in natural law. According to them, an absolute goodness, rightness, or justice, is given in principle. Those principal values ought to be respected as the source and foundation of the norms posited by human beings; only then those norms are just and therefore objectively valid. (The most well-known development of this idea is of course Plato's theory of the idea of the good.)
To this thesis, Kelsen objects that experience shows that even if there are such objective and absolute values, men clearly do not know them, nor act according to them. Men dispute about norms, men act unjustly, men do wrong. As he puts it:
"But on grounds of experience it will have to be accepted, that insight into the [consequences] prescribed by the order of natural law, and that knowledge of the 'right' or 'just' conduct is lacking in men, or at least in the majority of men..." .
Now the natural law-believer may come up with the solution Plato already thought of. It may be true that most men lack insight in the absolute values, but that does not mean that this insight is impossible in principle. There even may be in fact some men, or one person, who sees clearly what true justice or real goodness in itself is and demands for human life. Then only those men, or that person, ought to rule. For Plato, only the philosopher ought to be king.
But Kelsen points out that this solution means the authorization of a monarchy or aristocracy, whereas the claimed foundation of that authority is totally untestable for those submitted to it. For them, the question of the foundation of the norms posited by those in command, is not answered by referring to the natural good or just. For them, the foundation has to be found elsewhere.
Here Kelsen introduces the concept of the basic norm . Its status is that of a presupposition for a certain interpretation: to assume the basic norm, means to regard norms as objectively valid. Its function is to establish the validity of norms, or, to validate the normativity of norms as its objective meaning. Or, in terms of the author of norms, its function is to interprete the person who commands as authorized, legitimate.
However, Kelsen adds that basic norm does not have to be presupposed. One may or may not assume that norms are objectively valid. And this addition shows that the concept is morally problematic, even morally unreasonable. For first of all, the basic norm in itself offers no reason for the objective validity of norms. Its meaning is nothing but this: if a basic norm is presupposed, then norms are seen as objectively binding. Secondly, there is no reason at all to assume the basic norm. It is a matter of pure choice for the individual to assume or not the basic norm. Those who do not presuppose the basic norm, regard norms as acts of the will that are enforced with more or less psychic or physical coercion. According to Kelsen's analysis then, the validity of norms is a matter of arbitrary choice for the majority of men. Moreover, every reality except human acts, is a matter of complete indifference to them.
But eventually Kelsen's science of norms leads to indifference towards human acts as well, as will be shown next.

Second separation of 'is' from 'ought'
The 'is-ought' separation criterium not only requires that norms are not deduced from facts. It also requires that facts are not deduced from norms. The way in which Kelsen tries to meet this condition, can be explained by discussing his criticism of common language.
According to Kelsen the usual expressions "something ought to be, you ought to do this" lead to the idea that norms point to actual behaviour, that norms intend that a particular act takes place and that norms motivate behaviour. Also, one usually speaks about norms as if they create worthiness as a real attribute of persons. For example, the norm 'thiefs ought to be blamed' is expressed as 'thiefs are blameworthy', and this expression is understood as if blameworthiness is an objective attribute of the thief. Common language thus leads to the idea that norms imply facts, and lend values to persons and things. Of course, this idea then brings along a certain variation of moral objectivism.
However, this idea is a misconception according to Kelsen . His arguments to show this start with the truism that not the norms intend or will something, but the person who states the norm. What is it then, that this person intends? At first, Kelsen holds that the norm-stating person intends a particular fact, namely behaviour according to the norm and such that it is motivated by the idea of the norm. It follows then that a reaction is to be distinguished as norm-conforming behaviour from other reactions, if it agrees with the meaning of the act of the will of the person stating the norm, and if it is motivated by the intention to reach that agreement.
But the motivating role of the 'idea of a norm' is analyzed further by Kelsen, as follows. It can not be the case that the norm itself motivates, because a norm has no will, intends not. Neither can it be that the will of the commander motivates the behaviour of the addressed person, because that behaviour can only be motivated by the will of the addressed person himself. Therefore, the person stating a norm must intend that the other wants that behaviour himself, that he himself wills it as that which he ought to do . And only then his behaviour is motivated by acknowledgement of the norm.
This means that the intended acknowledgement of the norm can not be controlled by the one who commands. But neither is the actual behaviour a guarantee that the norm ruling this behaviour is acknowledged by the actor. Kelsen illustrates this point as follows . Take an officer commanding soldiers to shoot war-prisoners. It is possible that the officer utters this command while he wants that the soldiers ought not to shoot. In this case the will of the officer is ambivalent: his statement of the norm is no guarantee that he himself recognizes the norm. Moreover, it is possible that the soldiers obey out of fear for retaliation, while they disapprove of the command. In this case they act according to the norm without acknowledging it. From their standpoint, their behaviour can not be interpreted as a norm-observing act. In fact, all kinds of motives other than 'the idea of the norm' may come into play.
However, the personal rejection of, or disinterest in, a norm does not hinder others to understand and intend acts of will in a normative sense. It is quite possible that a person is held responsible for what he does according to norms that he himself does not agree to. But in such a case it is not possible to determine who is absolutely right: such judgement presupposes knowledge of an absolute norm, and, as we have seen, that presupposition can not be made rationally according to Kelsen. One can merely state that as long as in general certain norms are confirmed in value-judgements like approvals, disapprovals, rewards and punishments, those norms are apparently seen as objectively valid. The possibility that someone is judged by norms he himself does not affirm, remains open . Now, if the personal acknowledgement of a norm is not necessary in order to judge behaviour according to that norm, we have to amend Kelsen's initial theses that the commanding person intends norm-motivated normconformity and that following a norm requires being motivated by the idea of that norm. In fact, all that matters is a correspondence between norms and acts. Indeed, Kelsen draws this conclusion himself. A further problem arises, though. How is norm-conform behaviour to be understood if not in the common sense? The usual understanding, as if conformity were a real attribute of the factual act, would lead to the idea that the dichotomy between is and ought is bridged in actual fact. And this idea is scientifically unacceptable according to Kelsen.
At this point Kelsen introduces the concept of the modally indifferent substratum . Normconform behaviour is actual behaviour of which the modally indifferent substratum is identical to, is similar to, the modally indifferent substratum of the behaviour that is stated in the norm as that which ought. In the case of an offense, the two substrata are not identical. Kelsen gives the following example. The norm: 'human beings ought not to lie' is offended when in fact a person A lies. That is: in this case the modally indifferent substratum "person A lying" is not identical to the modally indifferent substratum 'persons not lie' of the norm 'human beings ought not to lie' .
Clearly, Kelsen introduces this concept to denote an act independent of the normative or actual mode in which an act can be represented. The concept serves to make the relation between norms and facts intelligible without comparing 'is' and 'ought'; these are, following the dichotomy, uncomparable.
The concept then also serves to avoid the common misconception that value-judgments express real attributes of acts. One calls an act 'just', 'right', when it harmonizes with a norm. One talks as if factual acts incorporate the values of rightness and justice. Against this misconception, the concept of the modally indifferent substratum enables a rational account of value-judgements. Namely: a value-judgement 'act x is right' is the statement that the modally indifferent substratum of the factual act x is identical to the modally indifferent substratum of the act x stated in the norm. A value-judgement then is of the same order as any judgement stating that one object is identical to an other object; a value is this identity between substrata of 'ought' and 'is'.

Jonas versus Kelsen: Jonas' concept of normativity
The discussion of Kelsen's science of norms showed the arguments for and the consequences of giving up the idea of goodness as the origin or end for our norms. I argued that moral arbitrariness and indifference result from Kelsen's theory. Precisely these two characteristics are condemned by Hans Jonas.
Jonas opens his "Das Prinzip Verantwortung" with the statement that since we technically can destroy ourselves and maybe all life on earth, the time has come to recognize the objective value of the human existence and of life itself . Against the modern idea that science requires the separation of facts from values, Jonas argues that the constitution of being itself can be seen as the normative foundation for that recognition.
What makes the moral difference, what constitutes the difference between moral interest and moral indifference, is the insight that being itself is not indifferent to itself, that being affirms itself continuously against not-being. According to Jonas, natural life shows this quite clearly. Activities like animal behaviour, and organic processes like digestion, betray that teleology is at home in nature. From that we can learn that being is interested, at least in itself. The interest of being means that there is at least one fundamental value as a good in itself, which is objectively, with being itself, given to men. That fundamental objective value is the value of being itself: it means that being is absolutely better than non-being .
However, it is not till this value is acknowledged by humans that the normativity of being becomes realized morally. The human acknowledgment requires that a person takes the interest of being upon himself, that he knows himself to be responsible for being. Responsibility, then, is the human moral answer to the normative condition of being. Jonas illustrates his thesis with the example of the newborn child. Here, the mere breathing of the baby cries out an ought to the others: 'you ought to accept me'. Here, the being of a tiny little existent contains immanently and understandably an ought for others: here the 'ought' is really existent . This example also shows that Jonas does not mean by responsibility 'following nature', or 'respect for nature in all its organic processes'. He does not plead for some naturalistic ethics, but for a normative ontology. Being itself contains the good in itself. But 'being' develops, it stretches in time, it is present as the history of becoming and passing away, it is in the fugitive and vulnerable life. Human life is prior to all other forms of life, since it is only there that responsibility can take place. Therefore, responsibility as the moral answer to the goodness of being is first of all the care for human life. The newborn child, simply be being there, manifests the fragility of the good of being and thus presents responsibility to the good of being as the first human duty.
Jonas lays great emphasis on the political implications of this ethics of responsibility. He argues that only those who understand the first commandment of being itself ought to rule. And so Jonas ends with the kind of natural law theory that is the target of Kelsen's criticisms.