X hits on this document

PDF document

Ordered by the House of Commons to be printed 14th July 2004 - page 23 / 216





23 / 216



Intelligence, though it may not differ in type or, often, reliability from other forms of information used by governments, operates in a field of particular difficulty. By definition the data it is trying to provide have been deliberately concealed. Before the actual content of an intelligence report can be considered, the validity of the process which has led to its production must be confirmed. For imagery and signals intelligence this is not usually an issue, although even here the danger of deception must be considered. But for human intelligence the validation process is vital.


Human intelligence reports are usually available only at second-hand (for example, when the original informant talks to a case officer4 who interprets – often literally – his words to construct an intelligence report), and maybe third- or fourth-hand (the original informant talks to a friend, who more or less indirectly talks to a case officer). Documentary or other physical evidence is often more compelling than the best oral report5, and has the advantage of being more accessible to specialised examination, but is usually more difficult to acquire. Conventional oral reporting can be difficult enough if all in the chain understand the subject under discussion. When the topic is unfamiliar to one or more of the people involved, as can be the case when details of (say) nuclear weapons design are at issue, there is always the chance of misunderstanding. There is in such cases a considerable load on the case officer to be familiar with the subject-matter and sufficiently expert in explaining it. It need only be added that often those involved in providing intelligence may for one reason or another have deliberately mis-represented (or at least concealed) their true identities, their country of origin or their employment to their interlocutors6, to show how great is the need for careful evaluation of the validity of any information which eventually arrives.


The validation of a reporting chain requires both care and time, and can generally only be conducted by the agency responsible for collection. The process is informed by the operational side of the agency, but must include a separate auditing element, which can consider cases objectively and quite apart from their apparent intelligence value. Has the informant been properly quoted, all the way along the chain? Does he have credible access to the facts he claims to know? Does he have the right knowledge to understand what he claims to be reporting? Could he be under opposition control, or be being fed information? Is he fabricating? Can the bona fides, activities, movements or locations attributed to those involved in acquiring or transmitting a report be checked? Do we understand the motivations of those involved, their private agenda7, and hence the way in which their reports may be influenced by a desire to please or impress? How powerful is a wish for (in particular) financial reward? What, if any, distorting effect might such factors exert? Is there – at any stage – a deliberate intention to deceive? Generally speaking, the extent and depth of validation required will depend on the


4 5 6


An official responsible for handling and receiving reports from human intelligence sources. Such evidence is no more immune to deception or fabrication than is oral testimony, though of a different type. The ultimate in such deceptions is the classic ’double agent’, who is infiltrated into an espionage network to discover, misinform, expose or pervert it. We have been assured that SIS has for half a century been viscerally wary of emigre organisations. We return to this below in the context of Iraq.

Document info
Document views584
Page views584
Page last viewedTue Oct 25 05:32:25 UTC 2016