ICRC databases on international humanitarian law
  • Print page
Commentary - Art. 23. Part III : Captivity #Section II : Internment of prisoners of war #Chapter I : General observations
    ARTICLE 23. -- SECURITY OF PRISONERS


    GENERAL

    The problem of protecting prisoners of war from the hazards of military operations first arose during the First World War, as a result of the development of long-range artillery. At that time, the belligerents agreed not to establish depôts for prisoners less than thirty kilometers from the front, and the authors of the 1929 Convention included this rule in Article 7, paragraph 1 . It also happened during the First World War that the presence of prisoners in the fighting zone was used in order to ward off attacks by the enemy artillery. Such practices were subsequently prohibited by Article 9, paragraph 4 , of the 1929 Convention.
    With the technical development of weapons during the Second World War the 1929 text became inadequate. The situation resulting [p.187] from the operation of aircraft over the entire enemy territory could be met only by a very broad interpretation of Article 7 and Article 9, paragraph 4 , of the Convention (1). The International Committee of the Red Cross therefore proposed to the Conference of Government Experts that the text should be improved, not only by broadening the scope of Article 9, paragraph 4 , but also -- still more important -- by introducing new provisions concerning air-raid shelters and protection from air bombardment, and the notification and marking of prisoner-of-war camps.

    PARAGRAPH 1. -- GEOGRAPHICAL SITUATION OF
    PLACES OF INTERNMENT

    This paragraph contains two distinct ideas; the first concerns the presence of prisoners of war in areas exposed to fire, and the second relates to the use of prisoners for purposes of protection.
    The first part of this paragraph, which relates to the detention of prisoners of war in areas exposed to fire, should be considered in conjunction with Article 19, paragraph 1 , which stipulates that prisoners of war must be evacuated as soon as possible after their capture to camps situated in an area far enough from the combat zone for them to be out of danger. Account must also be taken, however, of Article 47, paragraph 2 , which states: "If the combat zone draws closer to a camp, the prisoners of war in the said camp shall not be transferred unless their transfer can be carried out in adequate conditions of safety, or if they are exposed to greater risks by remaining on the spot than by being transferred."
    The words "at any time" make this provision absolute; regardless whether or not a truce or armistice has been concluded, prisoners of war must not be kept in areas in which the arms or armed forces of the two parties are in operation. The word "areas" must be interpreted very broadly and must not be limited to the range of weapons placed in any particular sector. Prisoners of war must be evacuated not only following capture (in accordance with Article 19 ), but also whenever any shift in the front may result in prisoners being in the combat zone.
    The stipulation that prisoners of war must not be used for purposes of protection implies that they must be interned away from any military objective; a distinction must nevertheless be made between [p.188] military objectives as such (troop concentrations, airfields, antiaircraft batteries, munition dumps, radar and tracing stations, munition factories, heavy industries, marshalling yards etc.) and the other nerve centres of a country which, without being primarily of a military nature, nevertheless inevitably contribute, whether directly or indirectly, to the war potential of that country by virtue of their economic importance. Prisoners of war not employed in rural areas, i.e. those who are assigned to industrial work, are naturally quartered in the vicinity of the towns or cities in which those industries are located (2). On the other hand, however, prisoners of war may in no case be interned in the vicinity of military objectives in the strict sense of the term, since they would thus be exposed to heavy bombardment. This
    interpretation is moreover confirmed by the text of Article 50, paragraph 1 (b ), which provides that prisoners of war must not be required to work in the metallurgical, machinery and chemical industries.

    PARAGRAPH 2. -- AIR-RAID SHELTERS

    This paragraph was inserted by the Stockholm Conference. The requirement that prisoners of war must have shelters against air bombardment "to the same extent as the local civilian population" implies that either shelters must be supplied for prisoners of war in the same conditions as for the civilian population, or that the existing shelters must be available to prisoners of war and to the civilian population alike. This will generally be the case for prisoners of war employed in urban areas. Shelters will not be provided for labour detachments interned in rural areas if none are available for the rural population there. During the Second World War prisoners were sometimes obliged to remain in their quarters during air-raids, to continue working or to help with rescue operations (3). Such practices are henceforth prohibited by this provision, subject to the measures which may be required for the protection of the prisoners' own quarters. On this last point, the Convention is not very explicit as it does not specify to what extent
    prisoners may be required to engage in the [p.189] protection of their quarters. Since such measures are undertaken mainly in the interest of prisoners of war, one may logically consider that the Detaining Power can require them to take the same action as is required of its own armed forces in similar circumstances, it being understood that only measures of passive defence are involved. On this point there is however a slight difference between the French text, which is in the conditional tense, and the English, which is in the indicative and confirms our interpretation.
    The last sentence of the paragraph states that any other protective measure taken in favour of the population must also apply to prisoners of war. This, of course, refers to the local population, since identical measures are not taken in rural and urban areas. If civilian workers employed in a particular industry are issued with special equipment for use during air-raids (gas masks, protective clothing, etc.), such equipment must also be made available to prisoners of war.

    PARAGRAPH 3. -- NOTIFICATION OF GEOGRAPHICAL LOCATION OF CAMPS

    In the course of the Second World War, the International Committee of the Red Cross made several appeals to the belligerents to notify the geographical location of prisoner-of-war camps, in order to avoid air-raids on them; this action was based on Article 8, paragraph 1 , of the 1929 Convention. The belligerents refused to comply with this request, however, on the ground that such notification might give the enemy valuable information regarding the location of certain important industries and that, moreover, the enemy might induce prisoners of war to revolt by parachuting weapons to them at an appropriate moment. In actual fact, the location of the main prisoner-of-war camps was known to the belligerents and the present provision was approved without difficulty at the 1949 Diplomatic Conference. It is understood that this requirement applies only to camps of some importance or base camps, and does not cover labour detachments unless they are fairly large and of some degree of permanence. In accordance with the spirit of the first paragraph of
    the present Article, the Detaining Power must not take advantage of the temporary presence of a few prisoners at a given place in order to afford protection to military objectives in the vicinity. It may also be that a prisoner-of-war camp originally situated far from any military objectives may, with changing circumstances, find itself in their vicinity. In the interests of security, it is essential to move the camp in good time.
    The risk of error, and the consequent danger for prisoners, will be greatly reduced if the belligerents notify each other of the geographical [p.190] location of camps in sufficient detail. What are the implications of the phrase "all useful information" in the Article? Does it mean a general indication of the district or, on the other hand, should sufficient details be given to enable the camp to be pin-pointed on a map? Only the latter information would be of any real use in order to keep the camp free from attacks. It is therefore recommended that belligerents should give all possible details, despite the objections to which we have referred above.

    PARAGRAPH 4. -- MARKING OF CAMPS

    At the beginning of the Second World War, the International Committee suggested to the belligerents that places of internment of prisoners of war should be marked in a special manner so that they could be clearly distinguished by enemy aircraft. This request was not generally complied with, as the belligerents feared that camps so marked would provide landmarks for aircraft, particularly at night (4). However, with or without the permission of camp commanders, prisoners in some camps took the initiative of displaying markings in the daytime, by means of the letters PW or PG (prisoners of war or ' prisonniers de guerre '); this procedure was recognized by the 1949 Diplomatic Conference, subject to agreement on any other system.
    The stipulation that prisoner-of-war camps should be marked is subject to a reservation which, in our view, is so drafted as to permit an excessively restrictive interpretation. "Whenever military considerations permit..." should not be construed as meaning that camps may only be marked subject to special conditions to be determined solely by the Detaining power. Whenever circumstances permit, camps must be marked. The argument that the marking of camps might provide landmarks for low-flying aircraft no longer carries the same weight as in the past, since technical developments have made it possible for aircraft to find their way without ground visibility. It is therefore desirable, and in no way contrary to the interests of the belligerents, that prisoner-of-war camps should be marked at all times, day and night, although unfortunately the Convention does not specify this. Marking in this way would preclude the recurrence of events which occurred during the Second World War, when prisoner-of-war camps were bombed on several
    occasions (5).


    * (1) [(1) p.187] See ' Report of the International Committee of
    the Red Cross on its activities during the Second World
    War ', vol. I, pp. 305-319;

    (2) [(1) p.188] See ' Report on the Work of the Conference of
    Government Experts ', p. 133. The Conference did not
    support a proposal by the International Committee of the
    Red Cross that the establishment of prisoner-of-war camps
    near large urban centres should be prohibited;

    (3) [(2) p.188] See ' Report of the International Committee of
    the Red Cross on its activities during the Second World
    War ', vol. I, pp. 312-313;

    (4) [(1) p.190] See ' Final Record of the Diplomatic
    Conference of Geneva of 1949 ', Vol. II-A, p. 254;

    (5) [(2) p.190] See ' Report of the International Committee of
    the Red Cross on its activities during the Second World
    War ', Vol. I, pp. 313-319;