702 



BELL SYSTEM TECHNICAL JOURNAL 



On the other hand, if the language is not pure, but made up of a set of 

 pure components 



having different equivocation curves with the system, then the total dis- 

 tribution will usually be made up of a series of ridges. There will be one for 

 each Li weighted in accordance with its pi. The mean equivocation char- 

 acteristic will be a line somewhere in the midst of these ridges and may not 

 give a very complete picture of the situation. This is shown in Fig. 11. A 

 similar effect occurs if the system is not pure but made up of several systems 

 with different // curves. 



The effect of mixing pure languages which are near to one another in sta- 

 tistical structure is to increase the width of the ridge. Near the unicity 



i'-a 



p(H,Nr 



Fig. 11 — Distribution of equivocation with a mixed language L = \L\ ■{- \L>. 



point this tends to raise the mean equivocation, since equivocation cannot 

 become negative and the spreading is chiefly in the positive direction. We 

 expect, therefore, that in this region the calculations based on the random 

 cipher should be somewhat low. 



PART III 

 PRACTICAL SECRECY 



21. The Work Cila.racteristic 



After the unicity point has been passed in intercepted material there will 

 usually be a unique solution to the cryptogram. The problem of isolating 

 this single solution of high probability is the problem of cryptanalysis. In 

 the region before the unicity point we may say that the problem of crj^pt- 

 analysis is that of isolating all the possible solutions of high probability 

 (compared to the remainder) and determining their various probabilities. 



