Following these preliminary clarifications, the field lies open for a theoretical examination of the subject under discussion. What can be and should be deduced from the rationale that underlies private defence with regard to the duty to retreat?
One factor in this rationale—the factor of the autonomy of the individual person attacked—certainly does not support the duty to retreat. This duty has the poten- tial to harm the person’s autonomy, which includes, inter alia, his liberty to remain in the place where he is, to continue his legitimate activities, and not to retreat.
Thus, for example, in German law, where the rule that ‘Right must never yield to wrong’ is deeply rooted, there is no duty to retreat in ordinary cases.791In order to learn something from the other central elements of the rationale of private defence (and especially the guilt of the aggressor and the social-legal order) as they relate to the duty to retreat, it is first necessary to define the true nature of this duty.
What therefore is the substance of the duty to retreat? From which principle is it derived? Traditionally it is acceptable to view the duty to retreat as a particular expression of the requirement of necessity.792Several scholars have hinted that this concerns a mixed question of necessity and proportionality.793In my opinion, the primary focus of the duty to retreat is actually to be found in the requirement of proportionality and not in the requirement of necessity. I shall clarify my last assertion.
I shall first examine the dominant case in legal literature in discussions con- cerning the duty to retreat: the aggressor endangers the life of the person attacked (ie, an existential danger, such as the aggressor attempting to kill the person attacked) and two possible modes of action are available to the person attacked:
789It should be noted that the definition of ‘deadly force’ that is accepted—by the author too—is quite broad and is not limited to force whose expected result is death, but also includes, eg, force whose expected result is severe bodily harm. See greater elaboration on the definition of ‘deadly force’ in sec- tion 3.8.4 above.
790See, eg, Hall et al, n 582 above, at 439; Robinson (1984), n 37 above, vol 2 at 85.
791See, eg, Kremnitzer, n 10 above, at 207; Herrmann, n 342 above, at 754–55; Fletcher (1978), n 1 above, at 865. But compare to another description of German law: Eser, n 26 above, at 632.
792For a view of the duty to retreat as a direct derivation from the requirement of necessity see, eg, Ashworth, n 183 above, at 293; the well-known ruling of the Supreme Court of New Jersey in the case of State v Abbot36 NJ 63, 174 A 2d 881 (1961); Williams (1983), n 1 above, at 509.
793See Robinson (1984), n 37 above, vol 2 at 80 fn 31, 85 fn 58; Kremnitzer, n 10 above, at 179 fn 3.
the first—the use of deadly defensive force against the aggressor (more moderate force will not be effective in the circumstances of this case), and the second—safe retreat from the scene of the event. The classic assertion of the proponents of the duty to retreat is that the requirement of necessity impels retreat under these cir- cumstances, for if the attacked person safely retreats, he is no longer in danger.
Consequently, the use of defensive force against the aggressor would be rendered unnecessary.
The counter-claim of those who reject the duty to retreat is that if the attacked person retreats, although his life and bodily integrity will no longer be at risk, other important interests will be harmed. The classic argument is that the retreat of the person attacked will harm his honour, since he will be portrayed as a coward. An additional argument that relates to the interests of the person attacked is that the duty to retreat will infringe his freedom of action and even endanger him physic- ally. Another—more modern and stronger—argument is that the retreat of the person attacked harms the social-legal order, because it constitutes a retreat of law and order before the aggressor who violates the law. Further on each of these arguments shall be examined separately, but at this stage the general picture will be presented.
My opinion is that despite the possibility of safe retreat, defensive force is nec- essary, and in contrast to the accepted argument, the requirement of necessity is actually fulfilled. However, although the interests necessitating defence do not include the life of the person attacked and his bodily integrity—since these can be saved by means of a retreat—they do include less central interests: the defence of the social-legal order, the freedom of action of the attacked person, and his hon- our. Consequently, the real question is no longer a question of necessity, but rather a question of proportionality: whether the use of deadly defensive force for the defence of the social-legal order, freedom of action of the attacked person and his honour meets the conditions of the proportionality requirement.
To illustrate the significance of this opinion, set forth below is Beale’s very inter- esting approach to the duty to retreat.794Beale concluded from the writings of Coke, that there was a duty in English common law to retreat from an aggressor who attempts to murder the attacked person, while no duty to retreat applied when the aggressor attempted to rob the person attacked. In the latter case, the person attacked was allowed to resist—even by use of deadly defensive force. The explanation that Beale gives for this state of affairs is based on the requirement of necessity: in a retreat from one who attempts to murder, the person attacked res- cues the attacked interest, ie, his life; while if the person attacked retreats from one who attempts to rob him (under the assumption that he is unable to retreat with the property) he will lose his property. A number of scholars, beginning with
794 In his famous and instructive article from the beginning of the 20th century ‘Retreat from a Murderous Assault’, n 40 above.
Foster,795 have attacked this odd opinion of Beale, and inter alia they have expressed the opinion that the latter did not correctly interpret the words of Coke796; according to their correct interpretation, the English common law did not demand a duty to retreat from a person who is attempting murder. In my opinion, Beale’s peculiar view stems directly from a perception of the issue of the retreat as a question of pure necessity and from overlooking the fact that it is actually a question of proportionality. If Beale had analysed the issue in light of the principle of proportionality, he might not have arrived at a result that actually permits the use of deadly defensive force for the defence of property.
An additional indication that the retreat should not be seen as a question of (mere) necessity, can also be found in the accepted view of moderate defensive force. As mentioned, it is accepted that retreat is not obligatory as an alternative to the use of non-lethal defensive force. If this was just a question of necessity, then exactly the same logic that obliges retreat prior to the use of deadly force (ie, the defensive force is unnecessary) also applies to a retreat prior to the use of moder- ate force.
In Section 3.8, I strove to base the flexible requirement of proportionality squarely on the rationale of private defence. Given my conclusion, that the ques- tion of the retreat constitutes a particular expression of the principle of propor- tionality, then the discussion of proportionality is very significant for the discussion of retreat as well.
A thorough examination of the issue of retreat shows, in my opinion, that the expression ‘duty to retreat’ is imprecise. The attacked person bears no duty to retreat. He is always entitled to remain in his place. Yet the deadly defensive force needed by the person attacked might not then comply with the requirement of proportionality, and thus he would be forbidden to use it. Clear evidence of the fact that, in effect, the duty to retreat is not a real duty, can be found in the fact that if the attacked person chooses not to retreat, and to suffer the attack without defending himself, there is no doubt that he does not breach any legal duty of any sort. And if the term ‘duty’ is nevertheless applied, it is necessary to be precise and to talk about a duty as a condition for the exercise of deadly defensive force and not about an independent duty. In effect, given the fact that attacked people rarely opt to suffer the attack instead of retreating, the difference between the duty to retreat and the prohibition of the use of defensive force is insignificant.
As mentioned, the classic argument of the opponents of the duty to retreat is that if the person attacked withdraws, his honour is significantly eroded. With
795See principally Perkins, n 40 above, at 139ff.
796Coke spoke of an attack whose purpose was ‘to rob ormurder’. Most of the scholars (beginning with Foster) understood his words as being alternatives, ie different cases (one—an attack with inten- tion to rob; the other—an attack with intention to murder), in both of which there is no duty to retreat.
Beale, however, interpreted the words of Coke as relating to a very specific type of attack, in which the aggressor presents the person attacked with the option of giving him the property and threatens that otherwise he will kill him—see Beale, n 40 above, at 573ff; Perkins, n 40 above, at 139ff.
regard to this assertion, significant differences exist among different cultures.
There were, and still are, cultures in which honour has an elevated status, with maximum acknowledgement of the person’s sense of honour, his stubborn will to stand up for himself, and his unwillingness to behave as a coward. The duty to retreat was, for example, traditionally negated in the southern and western states of the United States, while in the eastern and northern states a duty to retreat was imposed before the use of deadly force.797The accepted opinion today in legal lit- erature798is that the consideration of infringement of a person’s honour is only a secondary consideration, let alone it’s being of lesser value in comparison to the life of the aggressor. Beale described this point well, when he wrote already at the beginning of the 20th century:
So it is in the case of killing to avoid a stain on one’s honor. A really honorable man, a man of truly refined and elevated feeling, would perhaps always regret the apparent cowardice of a retreat, but he would regret ten times more, after the excitement of the contest was passed, the thought that he had the blood of a fellow-being on his hands. It is undoubtedly distasteful to retreat; but it is ten times more distasteful to kill.799; 800 However, the more modern reasoning for the negation of the duty to retreat is the injury caused by the retreat to the social-legal order. Feller writes that retreat from an aggressor implies the aggressor’s release from any danger involved in his aggressive action, the encouragement of aggression, waiver of right and a grant of supremacy to a breach of the law instead of guarding against it; in short, retreat from the attack sometimes means retreat from the law and public order, and not only retreat of the person attacked.801
Significance is also attributed to the deterrence of criminals that is achieved by the lack of a duty to retreat, and the ability of the aggressor to cease his attack.802 The factor of the social-legal order and also the ability of the aggressor to cease his attack create the difference that exists—with regard to the duty to retreat—
between private defence and the defence of necessity. While with regard to the
‘necessity’ there is no dispute that if a way to retreat is open before the actor he must use it (this conclusion is obligatory from the rigid requirements of necessity
797 See, eg, Beale, n 40 above, at 577.
798 See Perkins, n 40 above, at 160; Blackstone, n 40 above, vol 4, at 184–85; Kremnitzer, n 10 above, at 179 (maintaining that the injury to honour has only an indirect nature, since it concerns an interest that is not attacked and an injury that is not itself illegal).
799 See his article, n 40 above, at 581.
800 Perkins suggests a more technical reason for the negation of the argument that the duty to retreat will create a generation of cowards. As mentioned, even the supporters of the requirement for retreat restrict it to those cases in which it would be safe and would not endanger the retreating person who is under attack. Perkins notes that the classic possibility of safe retreat—in the rare cases where there is actually a possibility for safe retreat—exists when the attacked person has a firearm while the aggressor has a short-range weapon (such as a knife). In such a situation, when the aggressor is still not sufficiently close for his weapon to be effective, shooting him does not reflect heroism.
801 Feller, n 14 above, vol 2 at 429.
802 Eg, Feller, n 546 above, at 23.
and proportionality that apply to the necessity defence), with regard to private defence there is much dispute, and there are those who hold that the social-legal order dictates a negation of the duty to retreat.803Another view was expressed by Bein, according to which the value of the life and bodily integrity (of the aggres- sor) is more important than deterrence, especially since achieving deterrence is primarily the function of the state and is not imposed on the individual.804An even more far-reaching stance was expressed by Eser. According to his school of thought, viewing private defence as also fulfilling the social function of mainten- ance of social order actually implies the limitation of this right by means of the duty to retreat.805At first glance, this last position may seem odd. It can be under- stood, however, by perceiving the nature of the issue of retreat as a question of pro- portionality. Defensive force that seriously deviates from the requirement of proportionality not only fails to serve the social-legal order, but also even harms it significantly.806The factor of the social-legal order usually weighs significantly against the imposition of a duty to retreat. However, this is not a decisive weight in and of itself, but rather, requires the striking of a balance in which other factors will be taken into account. Some of these factors have already been noted and others will be noted below.
Another argument for the negation of the duty to retreat relates to the danger to the attacked person that is involved in the retreat. The main answer to this argu- ment is, as mentioned, that even those who support the duty to retreat do not demand its imposition unless the path of retreat does not endanger the person attacked.807An additional argument, which is connected to the argument of the danger, is that the rule of retreat suited the reality of the past—when the weapons were short-range (such as swords)—and that today, given the existence of firearms, there is no longer any room for a duty to retreat. The obvious answer to this argument is that even if the cases in which the option for safe retreat have become less frequent, there is still no room for an a priori negation of the rule that applies only to these cases (of safe retreat).808In addition, it is sufficient to review court judgments and to listen to, watch and read the mass media, in order to dis- cover that the use of non-incendiary weapons (other than firearms, ie, knives) is still prevalent today. Another argument is that even a retreat that appears to be
803See, eg, Kremnitzer, n 10 above, at 191. A number of scholars see the social-legal order as the explanation for the absence of a duty to retreat in German law—see, eg, Herrmann n 342 above, at 754–55. But compare with a different description of the German law in Eser, n 26 above, at 632.
804Bein, n 539 above, at 11.
805See Eser, n 26 above, at 632 (including n.55).
806See Figure 2 in Ch 1.6.3 above and the accompanying explanation.
807The requirement that the retreat should be safe constitutes a response to a significant part of the criticism that is directed against the rule of retreat; criticism that stems—in effect—from the incorrect application of the rule. See Hall et al, n 582 above, at 438 (the ruling State v Abbot36 NJ 63, 174 A 2d 881 (1961)).
808For discussions of these issues see Robinson (1984), n 37 above, vol 2 at 80; Perkins, n 40 above, at 151ff; Beale, n 40 above, at 578ff.
safe, usually bears a certain danger.809However, a case like this involves a relatively slight danger. Consequently, the risk cannot be a decisive factor, but it should be taken into account when assessing the balance for and against the duty to retreat.
An additional argument for the negation of the duty to retreat is the injury caused by such a duty to the freedom of action of the attacked person. We shall return to these arguments below.
The arguments in support of the rule of retreat usually focus on the importance of the value of human life and bodily integrity, and on the preference of this value for society as a whole above interests of relatively little value, such as the honour of the person attacked. Most of the writing that supports the duty to retreat is directed against the argument regarding the honour of the person attacked, an argument that was discussed earlier. This is because, in the past, this was the main argument of the opponents of the duty to retreat. The more modern argument is, as mentioned, the maintenance of the social-legal order. I believe that viewing the question of retreat as a particular expression of the principle of proportionality should enable the taking into account of all the values and all the considerations that were discussed above.
This leads to an examination of the treatment of the duty to retreat in Anglo- American law, in order to provide inspiration for an attempt to propose a suitable framework for this issue.