A Mistake with Lorenz

This continues previous articles about Lorenz. Over time the number of intercepted messages encoded with Lorenz increased.

A small number of 'depths' had been identified (where the same key fragment was reused on different messages), but there was not much progress being made in decoding the messages.

On the 30th August 1941. A German operator had a message to send. It was a long message, some 4000 characters. The operator (believed to be in Athens) correctly set up his machine, and proceeded to type a the long message. At the end of the message, the other operator who is believed to be in Vienna, sent a message which essentially said 'I didn't get that, please send it again'.

Rather than move to a new key, both operators put their machines back to the same starting position. This was forbidden, but both did it anyway. The Athens operator began to type the message all over again.

If he had typed exactly the same message then no harm would have been done to the German war effort, the allies would have only got two identical copies. Being human, however, he made slips, he abbreviated too as he must have been annoyed at having to type in the long message all over again.

The message began with the German phrase SPRUCHNUMMER, this means "message number" in English. The first time the operator typed in SPRUCHNUMMER. The second time he keyed in SPRUCHNR. The two encrypted messages were different from then on, though the plaintexts were essentially the same. Throughout the message the abbreviations meant that the second message was some 500 characters shorter.

The allies quickly realised importance of these messages, as the twelve letter 'message indicators' were identical, but the messages weren't, beyond the first few letters. The messages were sent to Bletchley Park and found their way to John Tiltman.

Tiltman combined the messages and was able to extract the original plaintext, and the string of key letters from the intercepts.

He now had what looked like a random string of key bits. He had no idea how these were generated, he had no clue how the machine was doing its job, but he at last knew what the machine was doing.

There are several threads of this story which remain.

We will need to discuss how the string of key bits was extracted from the message. We will need to discuss how the string of key characters was used to find the structure of the machine, and we'll need to discuss how this information was used to allow later messages to be decrypted quickly enough so that the information obtained was useful.