The Ethics of Spying 1
 

Normalization of espionage 3

The Lay of Dolon 3

International Law 9

Definition of Espionage 11

Spy vs. Traitor 14

Criminality of Espionage 21

Ruse de Guerre 32

Perfidy 34

What's wrong with perfidy? 43

Truthtelling 57

Conclusion 73
 
 
 

Recent highly publicized incidents of "penetration" of the United States intelligence "community" raise some troubling questions. The questions, however, are not of the usual sort, where we would want to find out who was responsible, how it happened, and how it can be prevented in the future. Instead I wonder how it is that we embarked on the normalization of the activity of spying in the first place. What is troubling about these cases [such as Aldrich Ames] is not that they happened, but that we don't expect them to happen now. After all, we won the Cold War, the Russians are now our uneasy allies, so why is it that they are still spying on us? Questions of who, how, and how must take a backseat to questions of why. In an attempt to answer this question, I would suggest that the end of the Cold War has put us in a situation where we can critically examine what was, under the necessity of inter-Super Power conflict, standard operating procedure.
 

If World War I was the application of industrial methods to the art of war, World War II saw the application of military methods to politics. I do not think that implications of the culmination of the idea of total war in that last global conflict are sufficiently appreciated. Total war is between entire nations, not just their respective armies. The legitimation of strategic bombing, the dropping of ordinance without any tactical purpose, is indicative of the shift. Even moreso, methods of warfare and inter-state relations which previously had been routinely denounced even while practiced, such as espionage, came to be legitimate as well. The shift is not all that difficult to detect,
 
 

Normalization of espionage

What should be surprising is that we accept the activity of spying as a normal function of government at all. Throughout most of Western history, spying has been condemned as unethical, contrary to the laws of nature, and punishable by death. The comment by *A and *B in the Harvard Journal of Law and Something to the effect that spying has not recieved the attention of ethical theorists to the same degree that war itself has is doubly disingenious, for they at once dismiss the longstanding condemnation of spying and suggest that there can be an ethics of spying. Parallel to this public condemnation has always been, however, the private recognition that spies are extremely useful, and thus indispensable to those in power. The bizarre status of espionage in international law reflects this ambivalence. Nations never (until rather recently) admitted to the use of spies, in spite of the fact that the employment of spies, in itself, has not been considered a crime (though it can make relations with the target state less than friendly). On the other hand, being a spy has always been a crime, and one punished with the harshest penalty. Even emerging international law of war in the past century only attempts to stipulate due process in the determining the guilt of spies, rather than questioning the traditional death penalty. What we will attempt here, then, is a recovery of the ethical standing of espionage within the overwhelming majority of the Western tradition of politics, with a view to an evaluation of the normalization of spying in recent times.
 

The Lay of Dolon

One of the earliest descriptions of espionage in the West is found in book 10 of the Iliad. It is worthwhile to consider closely the details of this depiction, since it is foundational in Western culture. The Trojans had forced the Greeks back to their ships, and both sides are uncertain about what will happen next, the Greeks fearing a night attack, wondering if they should not run away; the Trojans wondering if the Achaians are still ready to resist or are in rout and open to a final assault. When the Greeks take counsel, Nestor asks for volunteers,

'O my friends, is there no man who, trusting in the daring

of his own heart, would go among the high-hearted Trojans?

So he might catch some enemy, who straggled behind them,

or he might overhear some thing that the Trojans are saying,

what they deliberate among themselves, and whether they purpose

to stay where they are, close to the ships, or else to withdraw back

into their city, now that they have beaten the Achaians.' (10:204-210)
 

On the other side, following a parallel structure which will form the basis for our consideration of spying in the Iliad, the Trojans as well call a council, and Hektor speaks:

'Who would take upon him this work and bring it to fulfillment

for a huge price? The reward will be one that will suffice him;

for I will give a chariot and two strong-necked horses

who are the finest of all beside the fast ships of the Achaians

to him who has the daring, winning honor for himself also,

to go close to the swift-running ships and find out for us

whether the swift ships are guarded, as they were before this,

or whether now the Achaians who are beaten under our hands

are planning flight among themselves, and no longer are willing

to guard them by night, now that stark weariness has broken them.' (10: 303-312)
 

The first point to notice in comparing Nestor's and Hektor's proposals is the ostensible purpose of the mission. Nestor offers first the aim of capturing some stragglers, presumably for ransom or spoils, and only secondarily the aim of intelligence gathering. Hektor is straightforward in stating the goal of the mission is gathering needed information. A second point is the order of presentation. Nestor continues,

'Could a man learn this, and then come back again to us

unhurt, why huge and heaven-high would rise up his glory

among all people, and an excellent gift would befall him;

for all those who hold by the ships high power as princes,

of all these each one of them will give him a black sheep,

female, with a lamb beneath; there shall be no gift like this one,

one that will be forever by at the feasts and festivals.'
 

While the amount of speech that Nestor gives to the details of the information gathering and the rewards that would attend it suggest that the original goal he proposes in merely a pretense, the contrast with Hektor's speech is significant. Nestor only mentions reward as an apparent afterthought, as the spying is only incidental to seeking stragglers. Hektor begins with the reward, and only afterwards specifies what the task a hand entails. The nature of the two rewards is as well informative: Hektor offers what is not yet (and as it turns out, never is) his to give, spoil from the enemy which is dependent upon the success of the mission. Nestor suggests the Greek princes themselves give the reward, and the gift of livestock (for future festivals) only follows on the primary reward, glory among all people.

Next we have the volunteers for the mission. For the Greeks, Diomedes of the great war-cry steps forward, but asks for an accomplice and selects from those who offer themselves "godlike" Odysseus. The fact that more than one Greek is sent maintains the appearance that what is afoot is a raid rather than spying. Contrary to the descriptions of Diomedes and Odysseus at this point, the volunteer who answers Hektor's call is

. . .Dolon, Eumedes'

son, the sacred herald's, a man of much gold and much bronze.

He was an evil man to look on, but swift-footed;

moreover he was a single son among five sisters. (10: 314-317)
 

Dolon's first concern is not, like Diomedes, to secure a companion, but to make sure of the reward, and calls upon Hektor to swear on his sceptre to grant him the horses and chariot of Achilleus. Diomedes and Odysseus make no mention of reward.

Finally, we have parallel scenes in which the reconnoiters prepare themselves to set out. Diomedes and Odysseus arm themselves with swords, ( Odysseus in addition carries a bow), and put on helmets of leather (indeed, we are given the genealogy of the helm that Odysseus dons!). Dolon, on the other side, puts a bow across his shoulders and dresses himself in the skin of a grey wolf, and wears a cap of marten's hide, and takes a throwing spear. Euripides drama Rhesus, which deals with this episode in the Iliad, elaborates on Dolon's disguise.

Dolon:

I will fasten a wolf skin about my back, and o'er my head put

the brute's gaping jaws; then fitting its fore-feet to my hands and

its hind-feet to my legs I will go on all-fours in imitation of its

gait to puzzle the enemy when I approach their trenches and barriers

round the ships. But whenever I come to a deserted spot, on two feet

will I walk; such is the ruse I have decided on.
 

In the event, Dolon has the misfortune to encounter Diomedes and Odysseus, and being captured by them, is seized by the "green fear", begs to be ransomed, and on the reassurances of "resourceful" Odysseus, spills the beans. Dolon tells the pair of a recently arrived Thracian ally of the Trojans, camped apart from the Trojan army, whose prince possesses beautiful white horses. Having got this information, Diomedes tells Dolon not to think of escape.

'For if we let you get away, or set you free, later

you will come back again to the fast ships of the Achaians

either to spy on us once more, or to fight strongly with us.

But if, beaten down under my hands, you lose your life now,

then nevermore be an affliction upon the Argives.'
 

Dolon reaches for Diomedes chin to beg for mercy, but before he can do so Diomedes strikes off his head, which "still speaking dropped in the dust." Odysseus and Diomedes go on to raid the Thracians.

I would suggest there are four points in this section of the Iliad that are endemic to western civilization's notion of the spy. While every authority on war either admits or insists upon the centrality of intelligence, the means by which information is gathered can still be illicit. Spying, beginning with Dolon, is among those illicit means. But the reason for this ban on espionage is not at all clear. When we compare the parallel tales of Dolon on the one hand and Diomedes on the other we can see some of the features which make spying reprehensible. These are:

1. Motivation

2. Deception

3. Trustworthiness

4. The Death Penalty
 

Nothing is clear when it comes to spying. We may suspect that Dolon volunteered out of reasons less than noble. Of course, it would have been one thing if it was purely a matter of money, or gold and bronze. As it was, the fact that he had plenty of these does not change the fact that the condition of his service was some compensation. On the Greek side, one could say that the glory that would stem from the mission was foremost in the mind of Diomedes, and though this is different from monetary reward, it nonetheless is a form of self-interest that motivates the spies. But between the Greek and the Trojan the contrast is made, that one spy is only a spy conditional upon reward, and the other not.

Later theorists make a distinction on precisely these grounds. One who is a spy out of patriotism or loyalty is thought to be still honorable, whatever else may be the case. Thus we have honorable spies, properly motivated, and dishonorable spies, who are not. But while this may be a consideration, it is far from sufficient.

A spy is defined as a covert information gatherer. As in the case of Dolon, he relies on disguise and deception to achieve his end. Diomedes and Odysseus both dress as warriors, even though lightly armed (they do not carry shields, or ride in chariots). Dolon on the other hand pretends to be a wolf, not a Trojan warrior, and does not carry a sword. The use of a ruse is central here, and is what distinguishes a spy from a reconnoiter. A soldier gathering information will seek to be invisible, to escape detection, but he does not pretend to be something other than what he is in the process. A spy hides by pretending to be a civilian, or a member of the opposing force, or even part of the natural environs.⁽¹⁾ Again, as we see in the Iliad itself, this is a matter of degree that is only brought out by the contrast between Dolon and the two Greeks.

Third, when Dolon is captured, he holds back nothing, and even offers information in the attempt to save his life. This confirms our suspicion that he was undertaking the mission purely out of self-interest, so that when the situation changes he is equally capable of being a spy for the other side. No doubt Odysseus and Diomedes would have fought to the death if they were captured. But even so, we can never really know whether our spy is not doubled, or if we find that he is and triple him, that he is not quadrupled by the other side. The secrecy of the act entails that one completely trust one's spy, and at the same time is never able to do so. On the battlefield of course there are ways to prove one's loyalty beyond a doubt. The best we can do with a spy is to send along some quick acting poison just in case. And make sure they do not possess the latest intelligence on Tracian allies.

Finally, Dolon is executed. The killing of captives, especially wealthy ones, was not standard practice for Homeric warriors: the primary source of income besides the taking of spoils was the securing of ransom. The justification that Diomedes offers for not ransoming Dolon, however, would seem to apply to any ransom of a fighter, who would be expected to return to his duties on the side of the enemy. Indeed in the next book of the Iliad Agamemnon kills in battle two sons of Priam who previously had been captured by Achilleus and ransomed (11:101-121). So of the two reasons Diomedes expresses, the possibility that Dolon will spy again is the actual reason he is killed. The slaying of the two sons of Priam evokes sympathy and the sense of the tragedy of war. Dolon, it would appear, gets what a spy deserves.

[conclusion of this section]

International Law

The primary evidence we have for the illicit status of espionage in the Western tradition of international law is the admissibility of the death penalty for captured spies. In fact, this is about the only evidence we have. The admission of the death penalty for spying is paired with the recognition in several agreements on the rules for the conduct of war that espionage is an allowable means of war. Hugo Grotius, the 'father' of international law, says in his De iure Belli ac pacis that

. . . spies, whose sending is beyond doubt permitted by the law of nations-- such as the spies whom Moses sent out, or Joshua himself--if caught are usually treated most severely. 'It is customary,' says Appian, 'to kill spies.' Sometimes they are treated with justice by those who clearly have a just case for carrying on war; by others, however, they are dealt with in accordance with that impunity which the law of war accords. If any are to be found who refuse to make use of the help of spies, when it is offered to them, their refusal must be attributed to their loftiness of mind and confidence in their power to act openly, not to their view of what is just or unjust. (III.iv.3)
 

Thus we have something of a paradox in international law. On the one hand there is an unwillingness to explicitly categorize espionage as a crime or delict, and on the other a confirmation of the practice of nations in sentencing spies to death. This is strange in more than one aspect, the limitation of liability to the agent him/herself and not to the state for whom they worked going against a large part of international law. For example, the Nuremberg tribunals established beyond question that the defense of "superior orders" was not a defense at all where the order itself was in violation of international law. Lest we think this was an innovation of the post-WWII era, we should notice that spies have always been disallowed the "superior orders" defense, and this in spite of the fact that as far as most law on the matter was concerned, the orders to commit espionage were not, in point of fact, illegal. One would think that superior orders, where those orders were legal, would be an adequate defense. Since it has been held not to be so in the case of espionage, we should therefore conclude that the orders to commit espionage are in fact illegal under the rules of war.

There are several difficulties here, and we shall try to deal with them in some semblance of order. First we must come to an understanding of what constitutes espionage, according to the available law on the matter. Secondly we must try to determine what is the justification for the condoning of the death penalty for spying, and thus the rationale for holding it to be a crime under the rules of war. Thirdly, we will consider the ethical theory that is compatible with the second point, and see if it can justify at least parts of the current state of international law. Fourthly, we must then apply the ethical theory to actual practice, in order to see what can and cannot be justified under a consistent reading of both the ethical theory and international law.

Definition of Espionage

The Hague conventions, like most international law on the matter, are primarily concerned with restricting the crime of espionage, and thus its punishment, so as to guarantee the rights of other combatants. Thus we have a definition of what constitutes spying, intended to limit the use of the charge.
 

Art. 29. A person can only be considered a spy when, acting clandestinely or on false pretenses, he obtains or endeavours to obtain information in the zone of operations of a belligerent, with the intention of communicating it to the hostile party.
 

The General Orders 100 of the United States Army which served as a model for the Hague conventions, usually referred to as the Lieber Code after its author, specifies the crime without this concern, as is evident from the lack of a limiting adverb.
 

Art. 88. A spy is a person who secretly, in disguise or under false pretense, seeks information with the intention of communicating it to the enemy.
 

And unlike the Hague conventions, General Orders 100 goes on to specify the death penalty.
 

The spy is punishable with death by hanging by the neck, whether or not he succeed in obtaining the information or in conveying it to the enemy.
 

The Geneva Protocols again define espionage negatively:
 

3. A member of the armed forces of a Party to the conflict who is a resident of territory occupied by an adverse Party and who, on behalf of the Party on which he depends, gathers or attempts to gather information of military value within that territory shall not be considered as engaging in espionage unless he does so through an act of false pretenses or deliberately in a clandestine manner. Moreover, such a resident shall not lose his right to the status of prisoner of war and may not be treated as a spy unless he is captured while engaging in espionage.
 

The Uniform Code of Military Justice of the United States makes spying a capital offense, but without specifying precisely what the offense is.
 

Any person who in time of war is found lurking as a spy or acting as a spy in or about any place, vessel, or aircraft, within the control or jurisdiction of any of the armed forces, or in or about any shipyard, any manufacturing or industrial plant, or any other place or institution engaged in work in aid of the prosecution of the war by the Unites States, or elsewhere, shall be tried by a general court-martial or by a military commission and on conviction shall be punished by death.
 

From these definitions we are able to identify the pertinent features of espionage as a crime:

1. False pretenses, disguise or clandestine manner.

2. Intention to gather information.

(Success not necessary, but must be militarily important?)

3. Intention to convey information to hostile party.

4. Jurisdiction: act must take place within area controlled by target power.

5. Statute of limitation: spies must be caught in the act.

6. Death penalty is allowable.

And several other requirements seem presupposed, though not always explicitly stated:

1. The spy is a member of enemy armed forces.

2. There is a legally recognized state of war.

While the definition that emerges from international law is much more restricted than popular notions of spying, especially in the light of the Cold War, it has the advantage of being sufficiently precise that we may begin to analyze the nature of the crime and why death is accepted as a fitting penalty.
 
 
 
 
 
 
 
 

Spy vs. Traitor

The tradition in the West assumes that a spy is a member of the opposing military force. The crime involved in spying, then, is one of deception: a combatant either pretends not to be a combatant at all, or poses as one of the opposite side in the conflict, for the purpose of obtaining information. Intelligence gathering, insofar as it does not involve this sort of treacherous misrepresentation, is perfectly allowable. That is to say, a soldier captured behind enemy lines collecting information is to be treated as a combatant, which means he is to be granted quarter, and has rights as a prisoner of war. The gathering of publicly available information is of the same nature, unless, of course, deception is used in the process. The example given by the U.S. Intelligence community, in counter to the downing of Gary Powers, is a case in point. Two agents of the Soviet government obtained aerial photographs of the Glasgow Air Force base from the local Chamber of Commerce, and according to the tu quoque of the public report, falsely identified themselves as tourists.[ref. fas intell web page:] How could it be, though, that anyone would hand over sensitive information to a tourist, but not to any old tourist? The Powers affair is as well ambiguous, for one might argue that the flight, while attempting secrecy did not constititute spying, but as an act of reconnaisance was problematic enough, being a violation of the air space of a sovereign nation for purposes of gathering information when there was no state of war. As such, it was tantamount to an act of war. [best to leave this out, or footnote]

If the agent is not a member of the enemy military forces, the nature of the crime changes. If the person obtaining and conveying information to the enemy is in fact a member of our own military forces, or for that matter only of our government or even just an ordinary citizen, they are not, in the strict definition of the term, a spy. It may be the case that if a person turns over sensitive information to the enemy of their state, they have de facto become a member of the armed forces of the enemy state, but this oversimplifies the nature of the act. Handing over (traditio>traison>treason) one's country (or lesser social group) is a prima facie argument for the severing of ties to that entity--but not sufficient. Despite the oxymoronic sound of it, one could be a loyal traitor, if one retains one's allegience to one's own state while ostensibly aiding it's enemies.⁽²⁾ Treason could be the most severe form of civil disobedience; but here we would only be making again the distinction between noble and ignoble with reference to intention.⁽³⁾

Setting aside the "loyal traitor", however, the connection between treason and espionage becomes clearer: it is not the transfer of loyalty that consititutes treason, but the covert nature of that transfer, and the advantage taken of the deception of pretended continued allegience to the state one now works against. The point may be obfuscated by the rhetoric of solidarity often applied to politics today, and the difficulty of making an honest switch between sides. Where Clausewitz could easily transfer to the Russian military (and at least still fight the French) and back to the Prussian forces when that became feasible, Jomini who was a Swiss national served under Napoleon, but when he ran into difficulties went over to the Russian military, and then even returned to service in the French army. Presently such a transfer between warring military forces would be much more difficult, and a return almost impossible. We have defectors, rather than professional military officers, for the nature of conflict does not allow of exit. Much the same problem is faced by business corporations, where the relocation of an executive to a competing concern carries the threat that their knowledge of the internal workings and strategy of the company will provide an unfair advantage. But such a transfer is possible, so long as it meets minimal standards of openness and the often difficult maintenance of prior commitments to confidentiality, much as military officers in ages past could respect more than one allegience. The mere transfer of allegience does not constitute treason.⁽⁴⁾

As Judith Shklar writes,
 

Not all changes of political loyalty are betrayals, in any case. To change parties or convictions or even to emigrate may merely involve changes of mind.⁽⁵⁾
 

The transfer of loyalty that is motivated by benefits which stem from the transfer itself, rather than from the new loyalty itself, comes close to being treason. A mainland Chinese pilot landing his plane in Taiwan was at one time was offered a bounty of his weight in gold. Defectors often receive consideration from their new host government based not on any desert but purely on the fact that they defected and are a propaganda resource--an opportunity not offered to refugees from ostensible ally regimes. Now anyone who seriously considers emigration no doubt will have certain self-interested reasons for so doing, such as economic opportunity or better living conditions. When emigration entails benefits that stem from one's betrayal of one's former country (or corporation), such as intelligence or propaganda, and these benefits are what motivate the move, it becomes plausible that the act is perniciously selfish--there is no transfer of loyalty, for there was no loyalty to begin with. The defection must be secret, with all the character of an escape, for the other side seeks nothing other than the advantage to be gained from the defector's betrayal of his former allegience, and thus that state has a vested interest in preventing his going over.

The covert nature of defection implies that it involves deception, and this not only in the preparations for escape. If someone is of worth to an enemy nation, it is because they were invested with a certain amount of trust, which assumed their continued loyalty, which the person in question accepted. If they were sincere in accepting that trust, what could occasion the betrayal? A better offer certainly would not, for the person making the offer would realize that if the offer was accepted, no trust could possibly be placed in the traitor. No corporation would hire an executive who would divulge privileged information about their previous employer, precisely because they would know that there would be no qualms about betraying them as well at some later time. Defectors are dishonorable traitors.

The deception in treason is clearest when it is not just a matter of betraying a trust that was previously accepted after one exits, but of remaining within one's position in order to better combat one's former country. The best a defector can offer is intelligence which the victim state will know has been compromised, or some propaganda value; the traitor who stays in place becomes a valuable asset in a war. An ally who turns about on the battlefield, the commander who turns over his garrison, the saboteur who works at home, all these present opportunities that are only available with the use of traitors. And as well, intelligence that is reliable, up to date, and undetected by the target nations is much easier to obtain from traitors than from members of one's own military forces acting as spies. Thus traitors can function as spies while not meeting the requirements of the definition, that is, while not being members of the enemy armed forces.

The conflation and distinction of spies and traitors reveals much about what is taken to be the moral wrong connected with each. This is not a matter upon which there is consensus. Geoffrey B. Demarest differentiates the two actions on the basis of duty.
 

Spying and treason have a curious relationship. A key activity of traditional espionage is what amounts to the recruitment and development of traitors, and, although a traitor may also be spy, the traitor aspect will earn greater disrespect and loathing. The contempt accorded to the traitor results from the perceived breach of duty to one's country--a duty the foreign spy owes elsewhere. As can be seen, treason is somewhat different to espionage. ⁽⁶⁾
 

The Geneva Conventions (Art. 68) seeks to add to the requirements of the Hague Conventions that no one may be condemned to death for spying with out a trial, that no citizen of an occupied territory accused of spying may be sentenced on an unstated assumption of treason.

The death penalty may not be pronounced on a protected person unless the attention of the court has been particularly called to the fact that since the accused is not a national of the Occupying Power, he is not bound to it by any duty of allegiance.
 

Lieber's Code distinguishes between spies, who are members of the enemy armed forces; citizens of our own country who transmit legitimately obtained information to the enemy, who are traitors and should suffer death; and residents of occupied territory who inform the enemy (to whom they may owe allegience) who are war-traitors. The complexities of these distinctions becomes very clear when Lieber considers guides, who are necessarily recruited from the local population of enemy territory. Here we have a resident of occupied (or transited) territory who is made to give information to the occupying force, counter to his allegience to his own side in the war.

Art. 93. All armies in the field stand in need of guides, and impress them if they cannot obtain them otherwise.
 

The aspect of coercion here, though, releases the guide from being a traitor, and so from punishment.

Art. 94. No person having been forced by the enemy to serve as guide is punishable for having done so.
 
 
 

But if the guide volunteers for such duty, then the crime and penalty do apply.

Art. 95. If a citizen of a hostile and invaded district voluntarily serves as a guide to the enemy, or offers to do so, he is deemed a war-traitor, and shall suffer death.

Art. 96. A citizen serving voluntarily as a guide against his own country commits treason, and will be dealt with according to the law of his country.
 

And to make the position of the impressed guide even more difficult, he can also commit treason to the armed force which has coerced him, and suffer the penalty.

Art. 97. Guides, when it is clearly proved that they have misled intentionally, may be put to death.
 

The arena of conflicting loyalties we find here illustrates several points. Espionage and what Lieber terms "war-treason" are international crimes, according the the law of nations, or custom of war. Treason is only a crime of a particular nation, and reasonably so, for what is treason to one nation at war is loyalty to another. No doubt being a volunteer guide (or informant, etc.) for a occupying force is treason to one's country, not to mention fellow citizens, but we can hardly expect the occupying nation to regard it as a crime. Being a guide to one's own armed forces in one's own country that is occupied by an enemy force, on the contrary, is taken to be a war-crime, when we would be more likely to see that act as loyalty. Even misleading the enemy who has forced you to guide them seems to fall under this category. On this reading, treason is a lesser crime than violations of the laws of war, not in respect to penalty, which is death in almost all cases, but in jurisdiction: treason is not, and cannot be, a crime under international law.

Despite this difference, I think there is an underlying similarity between espionage and treason, one that explains the imposition of the ultimate penalty in both cases. The traitor and Lieber's "war-traitor" would seem to be very far apart, for one violates the duty and allegience owed his own country, while the other only violates the regulations of an occupying force (it would appear) to which she owes no loyalty. But the very difficulties we experience in explaining the criminality of espionage return here. We make use of traitors, and this is allowable, just as we would make use of "war-traitors" and spies, and yet the commission of these actions are punishable--in the person of the agent, not the command-- under national or international law.

Criminality of Espionage

What, then, are the grounds for the criminality of espionage? The singularity of a mandatory death penalty, as specified in the United States Uniform Code of Military Justice⁽⁷⁾, suggests a particularly heinous crime, if we can assume a proportionality between offense and punishment. Countries that have abjured the death penalty for all domestic crimes still reserve it for the crimes of treason and espionage.⁽⁸⁾ To the extent that the imposition of the death penalty is an expression of extreme disapproval of a crime, as in the denuciatory justification of punishment of Joel Feinberg*, the penalty for spying must represent an evaluation of the act itself, even while the denunciation is tempered by the willingness to use spies. What about spying is actually denouced?

The recognition of espionage as an admissible means in time of war argues for its utility. Unfortunately, for a particular means to be banned by the customs of war, it usually must be proved to also be militarily ineffectual as well as morally objectionable. Intelligence in general is indispensible for military operations, and espionage is one of the most effective methods of obtaining intelligence. Strategically, therefore, it behooves a general to not only use spies, but also to conceal as much as possible his own situation, and thus he must guard against enemy spies and traitors. Catching spies, however, is difficult business, for they are very well concealed. The prudent thing to do, then, is to deter espionage by assigning the ultimate penalty to it. As Emmerich Vattel writes in his The Law of Nations:

They are ordinarily punished with death, and justly so, since there is scarely any other means of protecting oneself from the harm they can do.
 

On this reading, it is the damage, and of course the attendant advantage to the enemy, that makes espionage a crime.

Deterrence is often invoked as a justification for punishment, both domestically and between nations. The allowal of reprisals for violations of the law of war is usually so regarded, not to mention the justification of standing armies and armament in times of peace, and even nuclear weapons. In the case of espionage there is a curious double-standard, however. Deterrence is usually applied toward actions that are banned, such as, for example, the execution of prisoners of war. The deterrent threat often, though not always, takes the form of the commission of the same act that it seeks to prevent. If the enemy kills our prisoners of war, we can respond in kind, proportionally, of course, and only to the extent necessary to convince our adversary to desist. The death penalty for spying, on the other hand, does not respond in kind (in that case the agent would turned, and spy for us), and the deterrence is not directed at the enemy but instead at the agent. We do not attempt to get our opponent to stop using espionage, but instead try to convince any possible spy that the risks are too high, and thus dry up the supply of spies.

There are, of course, very serious difficulties with a deterrence theory, whether for spies or for common criminals, the most telling of which is that deterrence is ineffectual. Where one might be able to make a counter-factual argument that punitive deterrence reduces the incidence of crime compared to what would have been the case in its absence, the amount of espionage does not bear the same relation to the harmful effects it produces: it is the amount of information lost to the enemy that counts, not the number of spies lurking about that causes damage. Thus a partial failure of deterence in reference to espionage can be tantamount to total failure. Besides, the quasi-ecomonic notion of the individual assumed by deterrence theory, the rational maximizer of pleasure envisioned by utilitarianism, may be completely inappropriate to espionage. The patriotic "honorable spy" will be more distraught at the failure of his mission than with death, and indeed he may carry a convenient means of suicide to limit the damage entailed by his capture, if we are to believe modern mythology on this matter. As for the "dishonorable spy," deterrence is up against a wall, for while one might raise the overall costs of illicit intelligence with the threat of a death sentence, those on the demand side of the curve can easily counter the disincentives with greater incentives, some of which may not be ignored.⁽⁹⁾ For these reasons, without having to resort to the history of espionage that has occurred in spite of the traditional penalty, deterrence cannot be a justification for the death penalty for spying, and thus the strategic harm that deterrence seeks to prevent cannot be the grounds for it as well.
 

If the laws of war were consistent on espionage, it would be as much of an offence to employ spies as it is to be one. The fact that it is not suggests the while states are concerned to avoid the damage caused by enemy spies, they are not ready to give up the advantages offered by employing their own. A cynic might suppose that this is why there can be a total ban on, again for example, the use of chemical weapons, the violation of which can be met by a counter-attack with the same: the advantage of using chemical weapons is small enough to surrender, provided the other side to a conflict does the same. But espionage cannot be banned: there is always the possibility that we may be able to gain intelligence that will be decisive through the use of spies, while countering whatever espionage is deployed against us. But then, again assuming consistency, there would be no crime in espionage, only a danger and threat to be defended against, like any other legitmate military tactic. But this is not what the recognition of the permissibility of the death penalty suggests.

The demand side of the equation deserves some consideration. If espionage is a crime, the employment of spies would also be a crime. Contrary to the denial of this by authorities cited above, it is certainly also true that most governments and military leaders are loath to admit that they employ spies. We might think that this is only a requirement of the secretive nature of espionage, where the admission of its existence may be the first step to exposure. But there is more to it than this. Even if it is admitted that everyone uses spies, the fact that it is a crime does taint the government using espionage, so that the admission is damaging not just in terms of the viability of secret services, but in terms of relations to other nations, even those one is at war with. Thus governments cannot legitimately, or that is overtly, engage in espionage. However, this is not purely a matter of political strategy or public relations. The employer of spies must keep a certain moral distance from the actual act.

The ethics of war often countenances moral gymnastics, or even contortions. A prime example is the "doctrine of double effect" attributed to Thomas Aquinas, that allows one to do what is otherwise not allowable if it is unavoidable. One can proceed with an attack upon the enemy even where this entails the forseen fatality of noncombatant persons, for example, so long as this consequence is not the desired effect or a means to some end. With spying the matter is more complex, for the crime is not merely a side-effect of the allowable activity of intelligence gathering, but is directly the means for it. The separation that makes espionage an allowable strategy, then, has to take place at a different locus. And this is between the employer and the spy.

The best strategic advice on the use of spies, from Sun Tzu to Frederick the Great, is to pay them well.⁽¹⁰⁾ Whether this is a recognition of the hazards and risks of being a spy, or the ease with which a spy might desert to the other side (and Frederick the Great used to drive his soldiers into battle by means of whips[ref for this?]), it does suggest that the only thing that can guarantee the loyalty of a spy is money. On the moral side, General Halleck thought that this cash nexus also created a break in the line of responsibility for espionage.

The employment of spies is no offense against the laws of war, and it gives to the enemy no cause of complaint. The matter is therefore narrowed down to simply a question of right between the general and the spy he employs. That the former may purchase and use the information so obtained we think there can be no doubt. Nor do we believe there is any wrong in his accepting the proffered services of a spy, provided the latter is made fully acquainted with the nature of his offense against the enemy and the ignominious punishment which must follow his capture. But we do not think a general is justifiable in seducing, by promise of a reward, a fellow-being to commit an act of treachery toward an enemy which makes him a felon at law and may subject him to a felon's death.⁽¹¹⁾
 

Thus the employment of a spy must be of the agent's free and informed will, which in a strict interpretion of Halleck's discussion, means that the general must wait for offers of services, the asking price of which will no doubt be sufficient to overcome the spy's estimates of the risks involved. Merely using such persons to one's own advantage does not entail any blame on the part of the commander, any more than taking advantage of a traitor or voluntary guide.⁽¹²⁾

There is an archaic feel to these condsiderations, however. The necessity of paying spies suggests that the motivation of the spy is personal enrichment, and Halleck's distancing of the spy suggests that the spy is an independent contractor rather than a partisan to the war. We are much more accustomed to think of spies as motivated by patriotism than Halleck allows. But in the past loyalty or patriotism was not thought of as a sufficient cause to personally engage in espionage. Vattel, immediately following the quote above, gives these reasons.

For this reason a man of honor, who does not want to run the risk of being put to death at the hands of a common executioner, refuses to act as a spy; moreover, he considers the office unworthy of him, since it generally involves some sort of treason.⁽¹³⁾
 

** Belli, in his A Treatise on Military Matters (date), pauses in his expositon of captial crimes in the military when he comes to desertion, and considers the question of whether a soldier could pretend to desert in order to spy.

Here arises a question about which I have often heard a discussion among soldiers of rank, namely, whether at the order of his general or his sovereign it is permissible for a nobleman to desert to the enemy, with the idea of spying out their situation and reporting it. Many judged this to be discreditable business for an honourable man of high standing; because both the act and the penalties are degrading; for anything is degrading which all good men so regard, inasmuch as ever what is honourable is determined either by law or by custom; again, an action brings no honour if it cannot be performed without loss of dignity.
 

As we might expect from the hypothetical tone of Belli's description of recieved notions of honor, he does not agree, and reverts to the honorable/dishonorable spy distinction to argue that it is permissible for a nobleman to be a spy.
 

Those upholding the other side of the argument declared that the action, as described, was not disgraceful, and that it ought not to be so classed; though the case is different with the low and base fellows who follow this pursuit as a business for pay, not, as the man in question, to assure the well-being of the state; for intent and purpose differentiate wrongs. Moreover, to imperil one's self for the safety of the public and in the service of the Emperor is a great honour.
 

We can establish a spectrum of positions, not necessarily on the honorablility of espionage per se, but on the permissibility of sponsoring it. Belli thinks it allowable for a superior to command an instance of spying, Vattel does not allow for an order to commit what for him is an unworthy act but does permit the enticement of spies.⁽¹⁴⁾ Halleck does not allow of either command or enticement, which means that one either responds to the asking price for spying or relies on the voluntary services of "noble" spies. This range of opinions could probably be completed with those who make no distinction between honorable and dishonorable spies, either Realpolitikers such a Machiavelli or Byndershoek who think any means are allowable in war, or Grotius' people of "loftiness of mind" who consider any form of espionage to be impermissible. Belli's argument against the dishonor of espionage, however, implies that this position was dominant, as we conclude also from the consensus on the law of war regarding espionage. But to argue that spying is dishonorable simply because of the ignoble penalty attached to it is to argue in a circle. We want to know why it should deserve the death penalty, which should also give an explanation of why spying is an affront to honor.
 

Vattel is most enlightening on this question. He argues, as does Halleck, that war does not mean the suspension of law, and so not all or any means are allowable in war. If spying fits ambiguously into the class of disallowed means, a statement of the critera for excluding certain means should allow us to answer the question of what is wrong with espionage. Vattel uses a variety of arguments to support the existence of law even in war, roughly falling into the categories of natural law, consequentialism, and legalism. All of these, however, have their basis in "faith between enemies".

Fidelity to promises and to treaties is the basis of the peace of Nations, as we have shown in a separate chapter. This good faith is sacred among men, and is absolutely essential to their common welfare. Are they to be dispensed from keeping it towards an enemy? It would be an error equally abhorrent and disastrous to imagine that all duties cease and all ties of humanity are broken when two Nations go to war. Because they are reduced to the necessity of taking up arms for the defense and the maintenance of their rights, men do not therefore cease to be men. They are still subject to the Laws of Nature; otherwise there would be no laws of war.
 

The idea that human beings are defined by sociability, as with Aristotle's zoon politikon, produces at once a basis for natural law in the necessary preconditions for social existence. If war is to be a human activity, it cannot undercut these conditions, whatever the benefits of doing so appear to be.

Secondly, the breaking of faith during war would have deleterious effects, for the control and conclusion of war requires the cooperation of both sides.

Now, the obligation of keeping faith, far from ceasing in time of war because of the precedence to be given duties towards oneself, becomes more necessary than ever. There are a thousand occasions during the actual course of the war when, for the sake of setting bounds to its fury and to the disasters which accompany it, the common interest and the welfare of both belligerents require that they be able to agree together on certain things. What would become of prisoners of war, of garrisons which capitulate and towns which surrender, if the word of an enemy could not be relied upon? War would degenerate into cruel and unrestrained violence, and there would be no limit to its calamities. How could it ever be brought to an end and peace re-established? If there were no longer any faith between enemies, the only certain end to a war would be the complete destruction of one of the parties.
 

These concerns are "humanitarian" in more than a single sense: not only is a world without good faith inhumane, but would as well be inhuman. On the other hand, these two grounds for the keeping of faith can be theoretically separated, and depending upon circumstance would produce different directives.

Thirdly, Vattel points out that the initiation of war abrogates only certain agreements between states, those treaties meant to regulate conduct in time of war cannot be dismissed, nor those actually made during the hostilities,

. . .for in treating with the enemy while he is actually engaged at war with him he necessarily, thought tacitly, renounces the power to break the agreement, . . . otherwise nothing would be gained by treating with the enemy and it would be absurd to make the attempt.
 

Along with the tacit renunciation of power to annul goes a "tacit condition of mutual observation," the essence of the contract. As with the consequentialist view of things, the legalist procedure has to point to the benefits that result from compliance, or the disutility that would result from general noncompliance. The Legalist only adds the specifics of what compliance, or breach, consists in, and so is concerned with procedure.

But Vattel goes further than this, to a more general characterization of good faith. From the necessity of being human, and the amelioration of conflict, and the sanctity of contracts, he produces the notion that good faith entails telling the truth.
 

Good faith does not consist solely in keeping our promises, but also in not decieving the enemy on those occasions when, for one reason or another, we are bound to speak the truth.
 

The constellation of evaluative terms surrounding good faith focus on a singular notion: truth-telling. As a matter of personal honor, keeping one's word is essential. In terms of consequences, the maintenance of good faith even in war is indispensible. And the legalist interpretation of when one's word must be given in good faith clarifies when and what the fulfillment of good faith entails. Vattel is mostly concerned with the latter, as we can see in his specification of what it means to lie. Lying only takes place, writes Vattel, "on those occasions when he is under obligation to speak the truth."

These principles having been laid down, it is not difficult to point out on what occasions it is necessary to speak the truth to the enemy and when it is lawful to lie to him. Whenever we have expressly or tacitly pledged ourselves to speak the truth to him, we are indispensibly bound to do so by that good faith the inviolability of which we have just set forth.
 

The entire purpose of the legalist specification of lying in bad faith is to distinguish it from allowable uses of deception under the laws of war. If we are not under explicit or tacit obligation to speak the truth, then lying or deception is not a violation either of natural law or customary law. Indeed, Vattel takes issue with "theologians" who hold truth to be something divine, to be safeguarded regardless of circumstances and consequences.

Of the foundations for the law of war regarding espionage, then, Vattel ultimately opts for consequentialist grounds. And as is often the case with consequentialism, by shifting the locus of moral evaluation from the act itself to some secondary value, the act becomes allowable under certain circumstances. If there is no obligation to speak the truth, and concealing or misrepresenting the truth produces a strategic advantage, then it is permissible.

But when, either by falsehood on occasions when we are not pledged to speak the truth or by some feint, we can deceive the enemy and gain an advantage in the war which it would be lawful to obtain by open force, there is no question that we may do so.
 

Deceptions practised upon the enemy, without perfidy, whether by word or act, and snares laid for him in the exercise of the rights of war, are strategems; their use has always been recognized as lawful, and has often constituted the glory of the greatest generals.
 

A further condition of the use of strategems is the properly humanitarian:

In the use of strategems we must respect not only the good faith due to an enemy, but also the rights of humanity, and we must be careful not to do anything which as a practice would be hurtful to mankind.
 

In addition to the strict obligation one might have to a particular party, there is also a responsibility to uphold good faith in general, much as one has a duty to tell the truth in trivial cases so that respect for truth and truth telling will prevail. We here have Vattel's principles for the use of strategems, the criteria by which we might distinguish those that are
 

Ruse de Guerre

The use of strategems gets a lot of attention in arm-chair military theory. But there is disagreement over the relative effectiveness of ruses. Clausewitz, for example, says that such things are of little strategic value.⁽¹⁵⁾ Vattel is quite concerned, as are many authors of his period, to overcome the classical Roman example which eschewed any form of deception. Thus Vattel, and those other authors, must be convinced of the strategic value of strategems or ruses. It is notable that Clausewitz does not discount ruses on the grounds of morality or international law, or even Roman tradition. The writers that do focus on such considerations are usually doing so only to argue against them. A position which settles between Clausewitz's dismissal and the traditional denunciation of deception has to assume not only that deception is permissible, but also that it can be effective. Thus Vattel, among others, assumes that a ruse can substititute for the direct utilization of force.

To a certain extent, the efficiency of the ruse presents us with an economy of conflict. If a ruse can substititute for force, then it is most valuable when we are lacking in force vis-a-vis our enemy, and to a lesser extent when we wish to conserve our forces. Repeatedly we see statements that historical examples only eschewed guile because they had at hand the necessary capacity to prevail in a straightforward manner. But the allowal of deception is not unfettered. If it is the case, in any particular use of a ruse de guerre, that the gains obtained by such means are countered by equal or greater losses in either the immediate or long-range future, then obviously the use of deception is counter-productive.

The two areas where deception is ruled out are, from Vattel's discussion, either where the particular benefits we gain from practicing good faith outweigh any advantage that might accrue from abusing it, such as in specific agreements about prisoner exchanges, et cetera; or where the overall results of the use of deception undercut the general practices of humanity so as to eventually result in making the world a much worse place, even for ourselves. Vattel gives the example of the use of distress signals to lure rescuers, so that they may be captured. Similar is the use of the flag of truce as a cover for an attack. Even where a ruse that takes advantage of such general practices of humanity produces immense benefits for our side in a conflict, we have to have very good reasons to justify violating the trust such practices, such as self-preservation, and even then the ruse may still not be morally justified.

There is a tension here between the economy of conflict and a tradition that holds deceit to be dishonorable. If deception is morally wrong, no amount of compensatory benefit can justify it. On the other hand, if all we need consider is enlightened self-interest, then there need be no prohibition which is overriden by the weight of more important considerations. But writers like Vattel and Grotius maintain an uneasy position between these clear cut alternatives. The retention of a restricted prohibition on deception, however, provides us with an insight into just what it is that makes such actions either disgraceful, illegal, or counter-productive. If we do not, as the noble Romans, eschew all use of deception, there still remains a class of ruses that are not permissible. These are either those that violate a pledge to speak the truth, or those which are harmful to humanity in general.
 

Perfidy
 

Cornelius Van Bynkershoek takes issue with the notion that there are limits in war in his work, Questions of Public Law, arguing that "in short everything is legitimate against an enemy."⁽¹⁶⁾ But even Bynkershoek has to admit some limits to the means that can be used in war, though he does so without giving up his initial position.
 

In my definition I was not even willing to omit 'fraud', since it is immaterial whether we employ strategy or courage against the enemy. Opinions differ, to be sure, and Grotius offers a great number of authorities and precedents on both sides. I would permit every kind of deceit with the sole exception of perfidy, and I make this exception not because of anything is illegitimate against an enemy, but because when an engagement has been made the enemy ceases to be an enemy as far as regards the engagement.
 

Bynkershoek has substituted the notion of "engagement" for Grotius' and Vattel's "good faith", which is significant in that it replaces the general requirement that one keep faith for the sake of humanity with a legalist interpretation that must specify an obligation owed to a specific party. But this is neither here nor there, for it exposes what is the objection to espionage, and thus the grounds for the impostion of the death penalty: perfidy.

A spy is someone who has "engagement" with the enemy. This is in fact what makes the spy valuable: the spy pretends to not be the enemy, and thus is privy to information which the opposition could not obtain by any other methods. But the spy uses this trust to betray those with whom he has engagement: he is given information on the confidence that he will keep it from the enemy, when in fact that is precisely his purpose. Here our earlier discussion of treason returns: the spy is the same as a traitor, except that the allegance he owes is feigned and not real.

The term "perfidy" is not often used of late. The origins of the word, however, help us understand why even such a political realist as Bynkershoek declines to allow it. Perfidy comes from the Latin word "perfidus", with the full phrase having been "per fidem decipere"⁽¹⁷⁾: to deceive by trust. The relation to a ruse de guerre is quite clear in that all ruses involve deception, but only certain ones are deception based on trust. The notion of "taking advantage," and that of "fair play" apply to**. The old saw, "fool me once, shame on you; fool me twice, shame on me" has application here. The responsibility for a deception lies with the deciever when the decieved had no reason to suspect such action, when she had trust in the the deciever. But once that trust is broken, the responsibility for a deception lies in the decieved's lack of caution and gullibility.

The breaking of trust, then, is the crucial point: we trust someone until they provide reasons for acting otherwise, but there is always the chance that those reasons will be revealed too late. For example, we assume that the person we meet on the street is not intent on doing us bodily harm. We walk alone and unarmed, possibly even offer a greeting. But when our assumption turns out to have been mistaken, we do not have readily available the means to defend ourselves. To the extent that our assailant has attempted to appear as a trustworthy fellow-citizen, until it was too late for us to take defensive measures, we have been decieved. So much more the case with the con artist, who pretends to be family, friend, or a government agent in order to gain our trust only in order to relieve us of our property. The worse case, however, is the person who, in times of war, pretends to be worthy of trust, and uses that confidence to betray us. The traitor or the spy are not so dangerous in what they do, however important the post they hand over or the information they convey, but in how they do it: per fidem.

Bynkershoek's notion of engagement, and Grotius' and Vattel's idea of good faith, have a general instance in those situations where trust must be presupposed. A distress signal ought to be sent in good faith, even though it is not necessarily directed at anyone in particular, and anyone who responds must also do so in good faith, without seeking to take advantage of the advertised distress. Both false distress signals, aimed at luring in those expecting to render aid rather than repel an attack, and false responses that intend to attack or rob those who expect aid, violate the trust that is entailed in the sending of such signals at all. Thus such actions are perfidy.

Certain conventions of war are of a similar nature. The flag of truce, like a distress signal, testifies that the bearer can be trusted to cease (at least temporarily) hostilities, and thus the other side must as well respect the the truce. No doubt this is what Bynkershoek has in mind when he speaks of engagement. The mutual (if tenuous) trust that attends the flag of truce is taken advantage of if one side violates the agreement for their own benefit. Such benefit is only to be obtained, however, because the other side believed that the truce would be respected. Again, they are not only decieved, but decieved by means of trust.

A further example from war conventions hints at a broader application of perfidy. The prime difficulty for theorists like Bynkershoek who define war negatively as the absence of all obligation, relation, and engagement with an enemy (Hobbes's state of nature being the root paradigm for this), is the identification of an enemy. Actually in theory this is no problem, for everyone who is not a friend is a foe. This assumes a primordial state of war, with peace being an artificial, if positive, human achievement. Those who are not our enemy must therefore have some distinguishing mark, so that we will know not to attack them and be able to relax our guard knowing they will not attack us.

Of course this is too simplistic. It reduces the world to only two categories, when it is obvious that there are degrees of hostitily and friendship between groups, and that within groups there are distinctions between those who are actively prosecuting war, those who are only passively involved or completely innocent, and even those that oppose it. The identification of an enemy is therefore positive, distinguishing the combatant from non-combatants, rather than friend from enemy. The distinctive mark, or the wearing of a uniform, or the open bearing of arms, is not meant to identify the soldier as a friendly, but instead is a declaration of hostility, or that is, a message to the enemy that we are prepared to fight. On this view, as Thomas Nagel says, hostility is a personal relation, and entails an "engagement" with the enemy: a declaration of war and the identification of oneself as an enemy. We can make this explict with examples of deception that play upon the convention of combatant self-identification.

Michael Walzer discusses the problems of combatant identity in chapter eleven of his Just and Unjust Wars. He relates an incident where a platoon of German soldiers in occupied France during WWII marched past what appeared to be a group of peasants digging potatoes in a field. At the proper time, the "peasants" dropped their shovels and picked up guns and fired. Walzer insists that the "peasants", who were actually French partisans, had an unfair advantage not only because they did not identify themselves as combatants, but also because they were citizens of a country that had officially surrendered. The German soldiers did not, and we might add, should not have, suspected that they would be attacked. As Walzer writes, "The surprise achieved by the partisans was of a kind vitually impossible in actual combat." The ambiguity we are examining is present here in Walzer's assessment of this incident: did the partisans have an "engagement" with the occupying force, which they violated by resuming armed conflict without giving notice, or did they take advantage of the immunity of non-combatants in general, and so resort to hostile action without identifying themselves as combatants? While this may seem a small point, it does bear directly on what was the unfair deception in this case, and thus on what about it may have qualified as perfidy, and thus what is wrong with spying.

If we hold, with Bynkershoek, that any means are permissible in war, except perfidy, then guerrila actions within occupied territory are permissible when there has been no surrender, and even the subterfuge of pretending to be civilian or non-combatant personnel is allowable. Even in an insurrection, where there may not be a publically recognized authority that conducts the war, and thus is capable of having "engagements" with the enemy, such methods are permissible. For if everyone who has not identified themselves as a friend is an enemy, then it is to be expected that we must treat everyone within an occupied territory (and indeed, without it as well) as an enemy. In order for guerrila resistance groups to practice perfidy, then, they would have to have identified themselves as friends, or at least as parties to an agreement of surrender.

The laws of war, however, do not take this approach. The Hague and Geneva conventions attempt to require that anyone who is a combatant in an armed conflict must so identify themselves. Typically this means the use of uniforms, but as this is not always practicable, the wearing of a distinguishing mark, such as an armband or cap is mandated, or at the very least the open bearing of arms. The French partisans in question, then, violated the rules of war by failing to identify themselves as combatants, which in this case is tantamount to falsely identifying themselves as non-combatants. The reason for making this requirement for combatants, however, remains unclear. On the one hand, the wearing of a distinctive mark provides for discriminate warfare, and makes possible the principle of non-combatant immunity. If it is the value of non-combatant immunity that motivates the convention, it is only in terms of that value that violations are to be judged. Using the cover of the civilian population in order to carry on a guerrila war has the disadvantage of making the civilian populace combatants as well, whether they will or no, but this is hardly perfidy unless we assume that non-combatant immunity is a prior agreement, on which the disguised guerrila depends in order to gain tactical advantage by violating it. Then indeed it is perfidy, and hardly requires Walzer's further specification of an agreement to surrender to render it so.

If we take for example a situation where an army has conquered a part of a territory, or taken a country without any agreed upon surrender, or for that matter an insurgency, we could ask whether the action Walzer considers would still be an example of perfidy. Here the partisans would have made no "engagement" with the enemy, and would only be violating some aspects of the Hague and Geneva conventions. So the question is, would such an ambush be a clever, albeit illegal, ruse de guerre or an example of perfidy? Certainly it would be a war-crime, given the existence of the conventions, but the international law on the question, unlike Bynkershoek, does not necessarily equate impermissibility with perfidy. Guerrila resistance often provokes reprisals, and with the nazis often disproportionate ones, which suggest that military personnel who have successfully defeated the military forces of a nation do expect that the non-combatants who remain in occupied territory to acquiesce to the new situation, and refrain from attacks upon the occupying forces or serving as guides to counter-invading forces, or conveying information to their government. Walzer's discounting of the traditional categories of international law, the "war-traitor" (the resistance patriot), or the war-rebel, while perhaps reflecting the actual practice of states, or more often insurrectionists, also betrays a shift in the location of perfidy. If perfidy only concerns explicit agreements with the enemy, then the wrongness of the potato-field ambush was in the fact that the attackers were Frenchmen who had ostensibly surrendered. But this assumes that everyone who has not declared themselves to be non-combatant is to be taken as an enemy: war is a state of nature.

The inverse of this view, and evidently the postition taken by international law on the question, is that everyone who does not explicitly identify themselves as a combatant is to be taken to be noncombatant. The status ante bellum here is taken to be peace, so that the failure to identify oneself as an enemy is to allow one to be taken for a friend, or at least a neutral. Now this means that a partisan pretending to be a peasant is relying upon the trust of his enemy, even where no specific engagement has been given, and thus such an attack is an example of perfidy, whether there had been a surrender or not. War treason makes sense under such an interpretation, for the treason involved in any such action is not a matter of going over to the other side, for one could quite plausibly assume that many residents of an occupied territory do in fact politically support the vanquished government, but instead the treason betraying the status, claimed by default, to be non-combatant.

There is a certain conventionality to military uniforms, and for that matter to the limitation of actual fighting to identified and recognized military forces, but these conventions rest on a particular concept of war as a relation between states, which therefore is amenable to regulation by international law. The notion that war is a relation, rather than the severing of all relations which Bynkershoek holds it to be, sets up certain requirements that cannot be avoided. The necessity of communicating hostile intent is the first of these, whether in a requirement for a declaration of war by the state, or in the wearing of distinctive marks or the open bearing of arms by an individual combatant. Failure to make such a self-identification does, in fact, produce some strategic advantage, as when a state declares war with a surprise attack, or as in the case of the potato diggers. But in either case the surprise rests on a tacit agreement, and thus is an unfair advantage, one that plays upon the trust of the enemy.

False distress signals, violation of the flag of truce, and disguise as noncombatants are examples of perfidy, in that they rely on the trust of the enemy in order to succeed. Strictly, therefore, they are not ruses de guerre which would be permissible under international law. Now it remains to be shown whether espionage falls into the same category.
 

We must accept that espionage is a strategem, one that is instrumental to the formulation of strategy and tactics that can only be based on reliable intelligence about the enemy. The question, again, is whether espionage is an allowable ruse de guerre or an example of perfidy. Again, it is not the potential advantage that is to be gained by espionage that makes it permissible or not, but whether it is consistent with the laws of war, and the moral position which underlies these laws. In other words, it is not whether information is obtained, but how it is obtained that will make espionage either a ruse de guerre or a war crime. This changes the focus of the illegality of being a spy (in contrast to the sending of spies), from a concern to deter potential spies to the delineation of what is of itself criminal. The central feature of espionage, as we can surmise from the definitions above, is the use of disguise or false pretences.

It is the disguise, or false pretense, which constitutes the perfidy, and forms the essential elements of the crime, which, by the laws of war, is punishable with an ignominious death.
 

Halleck, Id. 406:

The gravamen of the offense of a spy is the treachery of deception practiced, the being in disguise or acting under false pretense.
 

A spy, then, is someone who obtains information not merely through observation, nor even by stealth, but through misrepresentation: she is granted access to strategically important information because she pretends to be trustworthy. There are, of course, many ways of doing this, two types of which parallel our discussion above. First of all, the spy may make an "engagement" with the enemy, the most notable instance is the case mentioned by Belli, where a member of the enemy armed forces pretends to defect and become a member of our forces. He is then made privy (depending on the success of the ruse) to such information as he seeks to obtain based on the trust we now place in him as a member of our forces. Since he does this only to betray us, this is a clear example of perfidy. The lesser case is where a member of the enemy armed forces pretends to be a civilian, and on the basis of that misrepresentation is granted access to information that would not have been available to them otherwise. This may not be a case of directly promising fidelity, and so would not be considered perfidy under Bynkershoek's definition. But the failure to identify oneself as a partisan of the enemy is as much a deception, and if to a lesser extent still gives the spy opportunity to "lurk about" and obtain information that a self-identified member of the enemy armed forces would not have access to.⁽¹⁸⁾

Treachery, disguise, and false pretences are definitive of the crime of espionage. No doubt this is why Vattel and others maintain that no honorable gentleman would be a spy: such behavior is a violation of the code of honor. But as well, such actions are what entail the death penalty, as would other examples of perfidy. Even if we dissallow only perfidy in the conduct of war, as does Bynkershoek, we have to conclude that the commission of espionage is not allowable.
 

What's wrong with perfidy?

Establishing the essential nature of espionage as a form of perfidy may explain why the death penalty has be customary. But a full explication of the wrongness of espionage, in the moral as well as legal sense, requires that we explain what is wrong with perfidy. Much as is the case with treason, perfidy has been almost universally condemned. Most certainly it has not been held to be compatible with a European sense of honor. In a more consequentialist vein, perfidy has been seen as undermining the very basis of human society, the "good faith" of which Grotius and Vattel write. But either of these bases for the denunciation of perfidy are insufficient. Honor, what ever it may consist of, is culturally determined, and thus culturally specific. It is more a reflection of a widely held set of values than the origin of values. As Belli, citing Tiraqueau, writes:

. . .for anything is degrading which all good men so regard, inasmuch as even what is honorable is determined either by law or by custom.⁽¹⁹⁾
 

And when honor encounters higher values, when winning the game becomes more important than how it is played, an oversensitive sense of personal honor becomes a liability, and must be sacrificed to the greater good. The same could be said of the maintenance of "good faith". Indeed trust in all human relations is in itself a good, but in times of war the keeping of faith may become the perogative of the strong. Even the estimate of the damage inflicted on humanity, and thus the strength of the prohibition, by the use of perfidy will vary depending on whether one is winning or losing a war. Perhaps it will be the case that ending a war by the use of perfidy will be a greater good than the loss of good faith between humans. Or, as Vattel argues, the use of perfidy will make peace impossible, for the vanquished will not accept a defeat by treachery, and every and any war will not end until our enemy is made trustworthy by force, or that is, annihilated. But that cost may well be acceptable, if we think that the enemy has our annihilation in view, and attempts at negotiation, or even adherence to the laws of war will do nothing but make that task easier to accomplish.

Honor or consequentialism, therefore, are easily (and one might note, often) defeated by appeals to other values which allow for the use of any means against an enemy. We see the first eroded by the arguments of the early modern theorists of the law of war, who place emphasis on the consequences of perfidy rather than its affront honor. The allowance of espionage shifts the locus of honor from the act itself to the motivation for the act, so that the noble spy who acts in the cause of a just war commits no crime and only transgresses an antiquated ideal of honor, as no doubt Belli and Bynkershoek would maintain. The consequences for international society are then the only consideration that enters into international law on the question, and these are not so debilitating as to call for the outlawing of espionage, according to Vattel. But this leaves us more or less where we started, with the acquiesence to the practice of putting spies to death without an explanation of why the activity is different from other belligerent acts, which entitle the captured belligerent to the status of a prisoner of war.⁽²⁰⁾

I propose to take espionage as an instance of perfidy, and to show that this is the basis for the dishonor and the penalty attached to the activity. If the objectionable character of espionage lies in the fact that it is perfidy, the amount of damage done, either strategically or to humanity at large, does not factor into the reasons for holding it to be wrong. And if the potential damage which espionage can produce is not the basis for the acceptance of a death penalty for spies, neither can we allow for espionage on the grounds, as Vattel holds, that the damage can "to a certain extent be guarded against."⁽²¹⁾ But what of the damage that perfidy itself does? If spying is perfidious, the basis for disallowing perfidy must be made explicit.

We have already discussed the damage that may be done to humanity at large through the use of perfidy. But this does not rule it out absolutely. Good faith can take an awful lot of nibbling before it collapses, and perfidy can take place without producing a loss of good faith if it is not discovered.⁽²²⁾ Vattel asks whether is it lawful to recruit spies from the enemy or encourage treason, and concludes that it is.
 

We may take all possible means to weaken the enemy, provided that be not injurious to the common welfare of human society, as are poisoning and assassination. Now, in seducing a subject to act as a spy, or a governor to surrender his town, we do not attack the foundations of the common welfare and safety of mankind. The harm inflicted by subjects of the enemy, when acting as spies against him is not of a vital and unavoidable character, since it may to a certain extent be guarded against; and as to the security of fortresses, it is the business of the sovereign to select carefully those whom he puts in charge of them. Hence such practices are not contrary to the external law of Nations with respect to war, and the enemy is not justified in complaining of them as shameful outrages. Accordingly they are made use of in all wars. ⁽²³⁾
 

But when Vattel continues to consider whether the recruitment of spies and traitors is honorable, his conclusion is negative.
 

But are they honorable, and consistent with the dictates of a good conscience? Assuredly not; and this is the feeling of the generals themselves, since they never boast of having resorted to them. To engage a subject to betray his country, to bribe a traitor to set fire to a magazine, to tempt the fidelity of a governor, to corrupt him and lead him to deliver up the town confided to him, is to induce those persons to commit abominable crimes. Is it honorable to corrupt one's worst enemy and tempt him to crime?
 

Of course, as I have suggested, the dishonor of recruiting traitors is defeasible, since it is legally permissible, when it becomes a necessary strategem.
 

At best, such practices could only be excused in the most just war, when it would be a question of saving one's country from the destruction threatening it at the hands of an unjust conqueror. Under such circumstances the subject or general who should betray his prince would seem to be committing a less serious offence.
 

The conflict between loyalty and justice can indeed be a tragic one, and the crime of treason is only lessened by respect for justice. But it would seem that the recruitment of traitors is fully excused by a just cause. The acceptance of the proffered services of a traitor, according to Vattel who here is a precursor to Halleck's position, is not dishonorable in any case, as we may profit in his crime even while detesting it. Belli goes further to insist that being a spy is no dishonor, only dangerous, if undertaken from noble motives.

The range of allowance for espionage thus varies greatly. But it still is differentiated from those absolutely prohibited examples of perfidy, assassination and poison. Only Bynkershoek is willing to allow such actions in war, and does so in pointed contrast to Grotius. Clear examples of perfidy should make clear just what is the basis for the prohibition of perfidy, but as with espionage, if to a lesser degree, the international lawyers show little consensus.

The European bias against the use of poison is certainly pervasive, even today prompting agreements on the prohibition of stockpiling biological and chemical weapons where agreements on nuclear weapons are much harder to come by. The reasons for such a ban are varied. The just war tradition insists on discrimination in military action, and biological and chemical weapons are unable to tell the difference between combatant and noncombatant, or for that matter between friend and foe. From a tactical standpoint, this indiscriminate character of such weapons is only one aspect of the lack of control that is inherent in them, and for this reason we might agree to a ban since this is only to give up a weapon that is in fact impossible to employ efficiently. One might argue on humanitarian grounds that such weapons are excessively lethal and painful, inflicting unneccessary death and suffering. The traditional objection to poison, however, predates the development of either chemical or biological substances as weapons of mass destruction.⁽²⁴⁾ What is objectionable about these weapons is not lack of discrimination, for this has hardly been a restraining factor in the development of other weapons, but one of two features: the lack of defensive measures, and perfidy.

When Vattel says that spying is allowable because the victim is able to take some measures to guard against it, he harks back to the ancient notion of a fair fight. If war is taken to be a contest, a duel on a massive scale, it is essential that both sides adhere to a certain parity in weapons and tactics. If they do not, the contest settles nothing, since it is no contest. This Chivalrous ideal of fighting fair rules out not only the use of poison, but other forms of indirect attack as well, such as ambush, sniping, bombing, etc. The fact that a certain weapon or strategy cannot be defended against is not enough to define it as illicit, however. Certainly there is only a short lag between the development of an indefensible weapon and the the emergence of some, and often rather effective, countermeasures. But any weapon or strategy that attempts to produce a victory that is inevitable seeks to achieve an advantage that is not allowed to the other side. This is cheating, in that it attempts to fix the outcome of what is ostensibly a fair contest, to expose one's antagonist to danger without being exposed oneself.

One can take measures against poison, if one suspects that the enemy may be employing it. It may be extremely inconvenient to keep canaries and tasters on hand, to wear a sealed rubber suit, to go without water, but such things can be managed. It is not, then, just the unfairness of poison, as a form of attack against which there is no defense, that makes it despicable in the European tradition, or just the cowardice of such means, but the fact that it is conveyed under disguise, and in those things most intimate to us, what we ingest. Food, drink and air present themselves as necessities of life, and indeed they are; the poisoner uses this as the cover for an attack. Thus poisoning is a form of perfidy, where one uses the trust involved in living and breathing to betray the victim. This, I would suggest, is the grounds for the distate (so to speak) for poisoning in the Western tradition of the proper conduct of war.

Assassination has similiar features. Those theorists who hold that any action is allowable against an enemy usually stop short of recommending assassination. Again we might think that this is a reciprocal agreement between the heads of states, given that assassination is so difficult to guard against. But the defining aspect of assassination is not the danger it poses, but how it poses this danger. Vattel clarifies this by distinguishing between assassination and surprise, considering the classical case of an successful attack by Pepin, the father of Charlemagne, who crossed a river at night with a single soldier and killed his enemy in his bed. He holds this to have been a permissible attack, and that the use of torture or other penalties to deter such attack are only justified on grounds of self-defence, and has sharp words for any who say otherwise.

If anyone has absolutely condemned such bold strokes it was only done with the object of flattering those in high position who would wish to leave to soldiers and subordinates all the danger of war. ⁽²⁵⁾
 

Assassination is not only killing someone, an allowable act in war, but doing so in a manner which is unfair, destablising, and immoral: it is murder. In fact, the only distinguishing characteristic of assassination would seem to be that it is murder for a political cause.

How, then, do we distinguish between the surprise attack and assassination? The type of attack that would count as an assassination is one that it is difficult to defend against, but this is not so much a matter of the mode of attack as it is of the expectations of the victim. No doubt Pepin's enemy expected him to attack, and was only surprised by the time and place. But if for instance he was killed by one of his own trusted officers, the surprise would have been total, if no more fatal. This is the kind of surprise that is not possible in actual combat, an advantage that relies on the trust of its victim and the ability to betray that trust. What makes murder itself henious is not the fact of killing and dying, but that we have an expectation that our fellow human beings are to be trusted not to attempt to kill us; otherwise we would be in the Hobbesian state of nature and would trust no one and go about prepared to defend ourselves at any moment.⁽²⁶⁾ There are degrees of such trust, of course, depending upon social circumstances in any particular situation. But it is perhaps relied upon to the greatest extent in time of hostility, where one relies upon those who are expressly or tacitly allies. That someone would either use or feign allegience solely for the purpose of getting close enough to assassinate another human being relies upon the trust the victim has in his purported allies, and thus the surprise is not just a ruse, but perfidy in that it depends upon giving the victim some reasons to trust the assassin.

On this interpretation of assassination, it is not a completely different kind of crime, but only a murder that is complicated by a political conflict, and easily confused with a surprise attack that is allowable under the rules of war. Even an "ordinary" murder is to a degree an act of perfidy, for it is committed by someone who has not identified themselves as an enemy, and thus takes advantage of the tacit "engagement" we have with our fellow citizens.⁽²⁷⁾
 

Kant states several "Preliminary Articles for Perpetual Peace Among Nations" in his essay Perpetual Peace. The sixth article concerns us here.

No nation at war with another shall permit such acts of war as shall make mutual trust impossible during some future time of peace: Such acts include the use of Assassins (percussores) Poisoners (venefici) breach of surrender, instigation of treason (perduellio) in opposing the nation, etc.
 

Kant's point seems to be that certain strategems undermine the very possibility of peace, and taking seriously the central requirement of the medieval just war tradition-- that war must be for the sake of peace--such acts cannot be permitted.
 

These are dishonorable stratagems. Some level of trust in the enemy's way of thinking (Denkungsart) must be preserved even in the midst of war, for otherwise no peace could every be concluded and the hostilities would become a war of extermination (bellum internecinum).
 

Somewhat unfortunately, Kant here echoes the consequentialist arguments of the writers on the laws of war. But as we have seen, such reasons are not the only, or as I shall argue even the basic, reasons for not permitting "dishonorable stratagems". Kant's objection to perfidy appears to be that it would turn every conflict into a total one, the bellum internecinum, and given that we hold wars of extermination to be a bad thing, acts which lead to such conflicts must also be bad things. The starting point for any consequentialist theory has to be some posited value, and in order for the theory to be of any use there must be a consensus on those values. Wars of extermination may, in fact, be generally held to be bad things, especially if one is on the losing side. But the intrusion of other values, perhaps not commonly shared, vitiates this general consensus, so that a nation may well prefer to pursue a war of annihiation rather than suffer some more terrible fate. This is rather obvious in the modern era, where genocide has unfortunately become an everyday word, and where strategists have undertaken to "think the unthinkable" and developed "indefensible weapons".

As with the theorists of international law, Kant confuses the issue by mentioning eventualities. Certainly the elimination of the possibility of peace would be a bad thing, but is not necessarily the worst alternative imaginable on some schedules of value. Peace without honor may for some be a far worse fate than eternal warfare or even extinction. Vattel, for example, is constantly qualifying the deleterious effects of such strategems against the seriousness of the conflict, suggesting that in some cases peace without victory is less desireable than war without end. Observance of the laws of war, on a consequentialist reading, is only a matter of prudence, and prudence always answers to self-interest. That the self-interests of all the parties to war should coincide is something we could hardly count upon, a feature that is all the more disturbing since it is only in the interest of one if it is in the interests of all. The consequences of dishonorable acts, then, cannot ground an absolute prohibition, for they vary depending on the values held, the actual situation and the degree of reciprocity. Admittedly this is not a problem for those who hold that there is no such thing as an absolute rule in ethics, let alone in international law. But we still need to show how it can be that certain acts can be more nearly absolutely banned, or for that matter, why these acts should have such untoward consequences.
 

Treason and perfidy are unforgivable acts. The consequences that stem from their employment are not mechanical. If we follow the principle of the substitution of gile for force, we could say that the same result, say the capture of a stronghold, could equally be obtained either by the deployment of superior force or by the cultivation of a traitor. But the treason would create ill feeling that a purely military victory would not. Much the same can be said of other illicit means in war, from attacks on noncombatants to the use of chemical, biological, or nuclear weapons: a victory acheived by these methods is held to be unfair, to not be a victory at all. This implies that there can not be a substition of fraud for force in warfare, and that a purely instrumental understanding of the consequences of violations of the traditional laws of war may in fact make serious mistakes. But still the negative consequences of resorting to "dishonorable strategems" stem from their being held to be dishonorable or unethical, so that we first must establish the basis on which they are so, or else accept a strategic relativism that provides exceptions or justifies prudential secrecy.
 

We return here to a consideration of the difference between a ruse de guerre and perfidy. Poisoning and assassination necessarily fall under the category of perfidy, for they inherently rely on a deception that takes advantage of the trust of the victim. But espionage, in the looser use of the term, is not so clearly categorized. There is a confusion similar to that between the surprise attack and assassination in the case of espionage, where the allowable action (gathering information in spying, attacking in the surprise, or in general doing harm to the enemy) is aggregated with impermissible means of doing so. We must seek the line between a ruse and perfidy by examining how far deception can go without becoming perfidious, and through this show why perfidy is unethical.
 

Truthtelling

If we abjure the necessity of preserving good faith on consequentialist grounds, there still remains the notion of a duty to truthfulness. We might take this to be a prime value of honor, and possibly an archaic one. The ambiguity of the theoretical basis for good faith is highlighted by considering the direction of derivation of the moral rule of telling the truth. The consequentialist holds that it is better for me (or anyone) to live in a world where people can be trusted. ( The costs of verification, for example, are too high otherwise). From this we can then draw a general law that one should mostly tell the truth. Exceptions are easily granted where the case clearly demands the opposite for the maximization of utility. And finally from the general rule we can instantiate a particular rule that I should tell the truth to a particular person in a particular situation, given the proviso above. The duty to tell the truth, however, goes in the opposite direction, beginning with the rule that I should tell the truth to this person, and then generalizing to humanity at large. The necessity of truth-telling in a deontological ethical theory originates in the personal relation one has with the recipient of the communication, and only secondarily, if at all, takes into consideration the consequences of the act. This is not the usual interpretation of deontology, which takes duty to reside in a general law that has to be instantiated in a particular act, a general law that must be somehow rationally assumed or more simply divinely legislated. But this interpretation of duty leaves it as inexplicable as the notion of honor, while if we begin with the personal relation first, both duty and honor can be explained as ethical concepts without recourse to conventional values. To show this will require a good deal of argument.
 

Where consequentialism provides the general rule that it is better to tell the truth than not, is also allows for deviation from the rule where there are good reasons for doing so. The primary critique of deontology is that it does not allow for any exceptions to its rule, with the possibility of absurd consequences. Kant's categorical imperative is often so indicted, especially in regard to truth-telling. To always tell the truth would not only be socially unacceptable, but would also be a violation of other duties, as in the well known hypothetical case where the pursuing murder inquires as to the location of his victim, especially where that victim is someone to whom one has a duty of protection. But this is to misconstrue the nature of the categorical imperative, which is again perhaps due to Kant's infelicitous style. We can modify the strict formulation of a law of truth-telling into a prohibition of lying, and then further refine the duty by specifying that lying is only such to persons who have the right to expect truth from us. Now we have even more loopholes than the most radical consequentialist could produce.
 
 
 

In the first place, truth-telling is no simple act. In one instance it may mean the correct representation of objective facts, while in another it may means the communication of information we believe to be representative of objective facts. Further, truthfulness may mean the undisguised representation of subjective facts: sincerity. Again, telling the truth could be tantamount to not intentionally conveying false information, either by omission or by other means such as humor, inuendo, evasion, metaphor, etc. We might compose a continuum from factual truth to intentional truthfulness, and finally to only unintentional deception, all of which would fall on the truthtelling side of a dividing line.
 

agent	truthfulness	truthful	non- decept	non-decept	decept	decept
truth value	true	false	true	false	true	false
patient	gotit/not	not	gotit/not	not	got it	not

there seems to be a number of complications:

epistemological: the truth-value of the content of communication; scepticism comes into its own.

communicative success: regardless of content, this involves participatory activity, where the "if you know what I mean" has to be reciprocally validated, the content has to be acknowledged and then confirmed by re-transmission to the source. This further entails:

Participant status: the agent who is truthful seeks to communicate a content, and thus interested in its successful transmission. We must assume the patient to also have in interest in the same success. But in the case of the "non-deceptive" agent, it is the patient who demands communication, and the "agent" (who is more passive in this instance) does not have an interest in the transfer of information, and may very well have interests to the contrary. This is illustrated in the notion that where one cannot speak the truth, one must be silent. But as is often noted, silence itself is communication. Here the reciepent of information is the active party, and the one that must verify the information recieved, often in the face of a non-cooperating source. (the "agent and patient" characterization becomes less useful, as this role can shift; perhaps source and recipient of information would be preferable.) However, this is only a case of a previously hidden feature of normal communication being exposed by a certain pathology. In the ideal communicative situation (apologies to Habermas), the source has an interest in the correct transference of information, but so too does the recipient. The responsibility for verification is mutually initiated and conducted with a series of reiterations until both parties are satisfied of the correctness of the communication. The pathology of the single-sided inquiry of the recipient, such as Kant's example with the question "Do you have any money", is then mirrored with the single-sided interest of the source, again to be found in situations of unequal status. The teacher wants to be sure the student has understood, but the student is only interested in passing a course, so that the teacher cannot simply trust the responses of the pupil. A better example is in more explicit relations of domination, when a master seeks to convey a set of instructions to a slave, and then has an interest in those instructions being correctly understood. But the slave has no interest in understanding the instructions, for she will then be held liable for failure to comply with them.

Strategic communication, then, is that which resists unambiguous verification, and thwarts a single-sided attempt to convey information. But is this so different from the active attempt to convey false information? In terms of utility, undoubtedly it is not, if we place the utility of communication in the correct transmission of true information. Taken strategically, however, the utility is one-sided, where I have an interest in acertaining that the information I have is correct (meaning, practically reliable as a guide for action), while I have an equal interest in withholding correct information from my enemy, or even in providing him with false information. Stated simply, we are dealing with less than perfect communication situations, where the general utility of effective communication is replaced by the personal utility of effective disinformation.

Deception. The cases of the master and the robber present situations where communication is coerced by one side or the other. Both of these, however, still have an interest in obtaining the truth, either whether instructions have been understood or whether the victim in fact has any money. Responses by the unwilling partner to such communicative situations will seek to avoid conveying the truth, if it is not to their advantage (a not unlikely supposition, since it is worth the trouble of the coercer to attempt to extract it). Silence, or the possibly more effective equivocation, straightforwardly resist communicative success. This negative goal is evident it what Kant writes of equivocation.

If a man tries to extort the truth from us and we cannot tell it him and at the same time do not wish to lie, we are justified in resorting to equivocation in order to reduce him to silence and to put a stop to his questionings. If he is wise, he will leave it at that. ⁽²⁸⁾
 

Once, however, we undertake to avoid communicating the truth by communicating some substitute, we have shifted from the negative goal of escaping, as best we may, from the communicative situation to using it for our own strategic advantage. Here we cross a line.

The correlation of the ruse de guerre , lying, deception, and perfidy now can be made. Vattel and Kant both make a distinction between a false statement, a falsiloquium, and a lie, a mendacium. Now it would seem that the distinction hinges upon intent, that most slippery of moral principles. But although motivation may enter into the evaluation of non-truthtelling, it is not the distinction between a false statement and a lie. Now a falsiloquium necessarily involves the telling of an untruth, that is, the speaker knows her statement is false, or at least believes so. But this does not entail that the hearer take the statement as true. Much of language involves the use of expressions that considered strictly are patently false, from formal social interaction to tall tales, but almost no one is "taken in" by these, and then only accidentally. Deception would be the telling of an untruth with the intention of having it accepted as truth, and the aim of this differs from the use of falsehood for purposes of civility or entertainment in that it does not presuppose the willing "suspension of disbelief" on the part of the audience. Thus deception can be used for varying ends: pedogogical, as with the Santa Claus myth; medical, where patients are not told the truth when it is deemed bad for their health; or criminal, as in fraud. But deception is only a mendacium , according to Kant, if the auditor has the right to expect the truth. Thus it is not the truth of a statement that makes a lie, or even the intention of the speaker, but instead the right of the hearer to the truth.

How does someone have a right to the truth? Rights talk can be misleading, especially if taken in a possessive individualist sense. First we would have to determine who possesses a right to the truth, and under what circumstances they would be so entitled. But Kant and the theorists of the international laws of war hold that the right to truth is not only the possession of individual right bearers. The duty of truth telling to humanity at large can, it is true, be accounted for with a utilitarian analysis, but this would establish no rights on an individual basis, for it may well be that lying to any particular person at a particular time would in fact maximize overall utility. A natural or human right to freedom of information is more promising, but is hardly useful, as it is countered by a tentative rights to privacy, property rights, and by the practice of governments and individuals. Again we encounter the difficulties of beginning either with the status of individuals and proceeding to an aggregate, or with a general rule and proceeding to particular instantiations. But there is a much simpler method.

For Kant, anyone has a right to expect the truth from you to whom you have given reason to expect that you are telling the truth. Now while this does not seem any more definitive than the other attempts to establish a right to the truth that we have examined, it does have certain advantages. It certainly allows for instances when one does not tell the truth, in jest or out of consideration. It also releases one from any responsibility to tell the truth in situations where one does not volunteer to do so. And it does make it clear when and who has the right to the truth from us. In friendly interaction, it is often unclear whether one is, or should, be telling the truth. When I point out to my friend that his sock is untied, I am (except in the rare case of socks that require tying) telling an untruth, but of course from the nature of the statement he should know this, and if he is not paying attention, I've got him. This is the point of the game. But when matters of importance are at stake, the horseplay stops, and we must be truthful with our friends, for that is what makes them our friends. In the opposite case, if someone should seek to coerce us to reveal some truth, we have not given them any guarantee that what we infact tell them is true, and given the situation they must expect as much. Even the assertion that we are telling the truth, in such a situation, is insufficient, as the conditions for freely giving such an assertion have been violated at the outset. Thus any smart extractor of information will immediately resort to other means of verification, such as threats, bribery, or torture. If we give our word that what we say is true, even in a situation where we are hardly free to do so, and then lie, Kant says we are guilty of a crime, but not against the person we lie to, but against humanity at large.

We can contrast two very different situations here, that where someone tries to extract information from us and that where we wish to deceive someone else. Kant says that it is impermissible to tell a direct lie, for example, to the robber who demands to know whether or not we have money. Vattel, Grotius, Belli, and even Bynkershoek hold that it is impermissible to tell a direct lie to our enemy in order to deceive them, as in attacking under the cover of a flag of truce, or agreeing to an armistice or treaty we have no intention of keeping. In either case we seek to control the actions of others by giving them false information, and in at least some cases this may seem justified. A robber has no right to the truth from his victim, and those prosecuting an unjust war have no right to the truth from their victims, either. Yet even in such cases direct lying in considered impermissible. The law of necessity cannot override this prohibition, which leads to the absurditites that consequentialists use to deride Kant. But such an "absoluteness" to the prohibition on lying must lead us to understand the grounds for its absoluteness.

The allowable uses of deception are a clue to the basis for the disallowal of others. While lying in order to gain advantage may be prohibitied, for reasons we will come to later, the speaking or showing of untruths is not necessarily a lie. What makes this possible is that we have given no assertion to the intended victim of our deception that what we say or do is in fact true. Thus the conclusions our enemy draws from such information are strictly their responsibility. We may indeed present such information in such a way that would lead anyone to draw an erroneous inference, but as long as we stop short of personally guaranteeing such an interpretation, we are not responsible, and such deception is not perfidy. An example from ancient China may illustrate the point. In the days leading up to a crucial battle, Sun (Bin?) directed that the number of cooking fires in the army should decrease each night, in order that the enemy would think there were massive desertions and thus a lower troop strength.⁽²⁹⁾

Similarly, the Athenian generals Nicias and Demosthenes, at one point in the attempted flight from the disastrous siege of Syracuse, directed that cooking fires be lit prior to the army setting out on a night march.⁽³⁰⁾ And finally, Eisenhower (or Patton?) created the impression of multiple armies, and thus uncertainty over where the Allied invasion would land on the Continent, by maintaining radio traffic to and from the "ghost" divisions. In all these instances of ruse de guerre the deception relies on the enemy observing an action and being guided to an obvious, but false, interpretation of it. The communication that takes place in the course of a deception is indirect-- we do not communicate information directly to our enemy, but dissemble.

Kant allows of this sort of "lying". We can, according to Kant, make a false statement without saying anything.

But there can be falsiloquium even when people have no right to assume that we are expressing our thoughts. It is possible to deceive without making any statement whatever. I can make believe, make a demonstration from which others will draw the conclusion I want, though they have no right to expect that my action will express my real mind. In that case I have not lied to them, because I had not undertaken to express my mind. I may, for instance, wish people to think that I am off on a journey, and so I pack my luggage; people draw the conclusion I want them to draw; but others have no right to demand a declaration of my will from me. Thus the famous Law went on building so that people might not guess his intention to abscond.
 

On a par with the silent suggestion of intentions is the method of indirect communication, where we speak untruth to our conversational partner, who is "in on it" and so not decieved, and our conversation is overheard or intercepted by some third party.

Grotius' discussion of falsehood is enlightening in this regard. A falsehood, for Grotius, is a communication (by speech, writing, or signs) that necessarily differs from the thought of the sender. In order for a statement to be a lie, the speaker must believe that the statement is not true. But a stricter characterization is necessary to define a falsehood as a lie, and this is "a conflict with the existing and continuing right of him to whom the speech or sign is addressed; for it is sufficiently clear that no one lies to himself, however false his statement may be." Grotius is here in fundamental agreement with Kant, that the wrongness of a lie has to do with the rights of the victim. But the source of this right is further spelled out by Grotius.

By right in this connexion I do not mean every right without relation to the matter in question, but that which is peculiar to it and connected to it. Now that right is nothing else than the liberty of judgement which, as if by some tacit agreement, men who speak are understood to owe to those with whom they converse. For this is merely that mutual obligation which men had willed to introduce at the time when they determined to make use of speech and similar signs; for without such an obligation the invention of speech would have been void of result. ⁽³¹⁾
 

Whether the invention of speech entails some sort of original contract which requires truthfulness, or whether humans have a natural right to "liberty of judgment" we will put aside for the moment. It is the "peculiar" nature of the right which concerns us here, for this is what allows for exceptions to the impermissibility of lying.

There are four instances in which a right to the truth can be abrogated, according to Grotius. The first is when the recipient can not be wronged because they cannot possess such a right because they do not possess liberty of judgment, as Grotius says is the case with infants and the insane. (III.1.XII) The second, and the case which interests us here, is where the person to whom the falsehood is addressed is not decieved, or that is his liberty of judgment is not impinged, but a third party draws a false conclusion. The person addressed may not be decieved either by tacit understanding, as with figurative language, irony, hyperbole, or satire, or by arrangement, where for example two people act out a fantasy together. But for the third party, there is no deception, for as Grotius says,

There is no falsehood, again, in respect to him who chances to hear what is said; the conversation is not being held with him, consequently there is no obligation toward him. Indeed, if he forms for himself an opinion from what is said not to him, but to another, he has something which he can credit to himself, not to another. ⁽³²⁾
 

Grotius' third instance of permissible falsehood is where the "victim" more or less consents to being decieved. These are those "white lies" one gives to a sick friend, or to encourage one's fellow soldiers in a battle This is probably tricker to ascertain than in the case where the "victim" agrees to be lied to and is not decieved, but if there really is a desire to be decieved, a "liberty of judgement" is preserved. ⁽³³⁾

Grotius's fourth condition that permits falsehood is the superior right of the speaker over all the rights of the other. No doubt he has in mind the right of the master or sovereign over subjects, and lists suppliants, strangers and deserters as being in the class of persons without rights over those who receive them. But Grotius includes those "held by no bond of good faith," and immediately moves to the example of Pepin (cited above). The shift, or equation, of those held by relations of subordination and those outside of any relation (as opposed, I presume, to an equality of rights) illustrates a fundamental ambiguity in the laws of war. A just war presupposes a relation of superior right, and thus the permissibility of falsehood. Any means of deciet in war should thus be justified. But on the other hand it is obvious that there is no recognition of superior right, else there would not be a war. So far as concerns the conflict itself, then, there must be an equality of right, which implies a bond of good faith and thus the impermissibility of falsehood. We see, again, this point accepted by Grotius, who follows with an admission of the permittability of sending spies, but denying that of "assassins who act treacherously". Just who we owe the truth to, to whom we are bound by bonds of good faith, remains unclear.
 

The notion of a "right to truth", as mentioned earlier, is less than helpful. Of course, if the determination of such rights was simple, the domain of the application of a moral rule against lying would be unproblematic. But even the clear dichotomy between those who deserve the truth and those we can lie to is untenable. Bynkershoek's sudden transformation of an enemy, one who presumably has no rights, into a friend with the occurance of an engagement seeks to avoid blurring just such a sharp distinction. The prohibition on assassination and poison, on the other hand, suggests that there is no such dichotomy, and that even an enemy deserves a certain degree of good faith, and thus there is no class of persons that in Grotius' words possess liberty of judgment and nonetheless can be lied to.

Now it should be apparent that the sanctity of "liberty of judgment" for Grotius is an equivalent to one of Kant's formulations of the categorical imperative. Insofar as we respect another as an end-in-itself, we must also grant them their own autonomy. To give someone false information is to coerce them into making a decision, for by manipulating the conditions under which they must judge, we control their judgment. And in so doing, we use them for our own ends, treating them as a means for our purposes. Unless we intuitively agree that this is a bad thing, however, the immorality of this is not apparent, especially in the case of those others whose autonomy we have no reason to acknowledge. Grotius' willingness to supervene the right to liberty of judgment with superior rights is evidence that he does not share Kant's universalism, and thus the prohibition on lying assumes a specific character based on the relation one has with a communicative partner. In this Grotius is out of step with the more modern and cosmopolitan Kant, who maintains a more generic right to the truth that is not so readily defeasible. Rousseau objected to this feature of Grotius, after initial enthusiasm for his work, when he realized that Grotius held that a populace could agree to surrender its freedom, in perpetuity. The possession of a right to the truth, then, here splits between a rational construction which bases such a right on the ability to make autonomous judgements, and the legalist construal which limits "liberty of judgment" to those with prior recognized possession.

The rationalist approach presupposes at once both a prima facie granting of equality and respect to any being whom we encounter (even an enemy), and the insurmountable right to autonomy, and thus liberty of judgment, that any rational being necessarily entails. Kant gives us this in the third formulation of the categorical imperative: "For rational beings all stand under the law that each of them should treat himself and all others, never merely as a means, but always at the same time as an end in himself." Now respect for autonomy certainly means that we should not lie to another, for in so doing (as asserted above) we treat them as a means to our own ends. But it also rules out Grotius' fourth class of those who it is permissible to tell falsehoods to: there can be no right superior to the autonomy of a rational being. Any being that is rational cannot cease to be so by the act of a rational agent under Kant's theory of morality. For a rational agent to be justified in overriding the autonomy of another rational agent, that agent would have to universalize the maxim of the action as a universal law for all rational beings. But in so doing, the agent in this case would have to give equal right to any other rational being to override his or her own autonomy, including the freedom by which he or she decides to nullify the autonomy of the other. In simple, it is impossible to universalize superior r