The Siren Queen: An Ursula Blanchard Mystery at Queen Elizabeth I's (21 page)

Still, I hesitated, and Scrivener glanced at the Brockleys. “My companions are altogether reliable,” I said quickly. “They helped me to obtain the documents you hold in your hand.” I hesitated once more and then made up my mind. “I think I must be open with you. Cecil did vouch for you. God help us all if I’m making a mistake. You are right. Ridolfi wrote these letters.”

“Better men than Cecil have been deceived,” said Scrivener seriously. “But not by me. You needn’t fear to be frank with me. However, tell me only what I need to know. I won’t ask precisely how you came by these. They are not the originals, I take it?”

“No. They’re copies that we made. We were careful over accuracy.”

“I’m glad to hear it! Well . . . ” Scrivener laid the letters down and drew his writing set and a sheet of blank paper toward him. “We can but test a few possibilities. What do you know, already, of ciphers?”

“Very little. I have come across one, but that was something that a couple of brothers had invented privately and it wasn’t like this. It didn’t substitute letters or numbers for other letters—it used words for letters. This is quite strange to me.”

“I’d like to hear about the one you encountered. It sounds intriguing. However, as a start, let me show you what, in all innocence, two years ago, I showed Signor Ridolfi. The point is that in the most usual ciphers, the letters in the clear, ordinary text are replaced by code letters, numbers, or symbols. The cipher is harder to break if the system lets the replacement letters—or numbers or symbols—vary. I mean, you could have a code in which the letter A was always replaced by, say, W. But it would be better if A was sometimes represented by W and sometimes by some other letter. Perhaps by more than one other letter. Or numeral. That was what I had in mind when I invented this. Look.”

He began to write. I pulled a stool up close and watched. “First of all,” he said, “I put the alphabet across the page, like this . . . ”

a b c d e f g h i j k l m n o p q r s t u v w x y z

“. . . and then, beneath it, I write numbers, 1 to 26, since the alphabet has 26 letters in it. Like this, making sure that each number is neatly placed under the letter to which it refers. Thus . . . ”

“That’s the first stage,” he said. “It doesn’t stop there because if it did, it would be too easy to break—even if we complicated things a trifle by, say, moving all the figures along one or two places like this . . . ”

He wrote again, showing us what he meant. The result was:

“You could use any sequence of figures,” I said, interested. “I mean, you could start at—say—32, and go on from there, 32, 33, 34 . . . or begin at 105 . . . ”

“You could indeed,” said Scrivener, regarding me with approval, as a Latin master might consider a pupil who had grasped the ablative absolute at the first attempt. “But I didn’t complicate matters so much when I talked to Ridolfi. In fact, I just showed him how to work the code using 1 for A, and going on to 26. Now, on that basis, suppose we wanted to encipher, oh, shall we say your name? Ursula. That comes out as 21, 18, 19, 21, 12, 1. Any halfway competent clerk could break that, probably inside ten minutes, but, aha!”

He beamed at me. “Now we go on to stage two. We have turned your name into figures. On each of them, we perform a little arithmetical calculation, but not the same one every time.
The system I showed Ridolfi was that the first figure—in this case, 21—is multiplied by 3, the second by 4, the third by 5, the fourth by 6, and then you start again, multiplying the fifth figure—that’s 12—by 3—and so on. Are you accustomed to figurework at all? And can you use an abacus?”

“Yes. I used to help my uncle Herbert with his accounts,” I said.

Scrivener went to a chest, opened it, and brought out a box, from which he took a small abacus. It was a pretty thing, with wooden beads painted in brilliant colors. He handed it to me along with his quill pen. “Can you do it for your name? You multiply 21 by 3 . . . ”

I tackled the task, arriving after a while at 63, 72, 95, 126, 36, 4. “The letter U occurs twice in my name,” I said. “When we did the first stage of the code, U came out as 21. But now that we’ve done a calculation on each figure, the letter U has worked out to 63 and 126. Quite different.”

“Yes. The more different calculations you have, the less chance there is of a repetition. I found that four was the minimum if you really want to avoid that. If we’d only used three calculations, your letter U would have been 63 both times.”

I worked this out and found that he was right. The Brockleys had come to lean over the table as well. Brockley said: “It’s difficult to keep track of where you are.”

“In the sequence of calculations, you mean?” I said. “Yes, it is.”

Scrivener looked slightly pained. The pupil who had done so well with the ablative absolute had taken a severe toss at the gerund. “When you write out your first stage of numerals,” he said, taking his quill back from me, “you first of all write out the numbers you’re going to multiply them by—like this.” He wrote quickly.

3 4 5 6

“Then,” he said, “you put the enciphered numbers, produced by your first stage, beneath them in columns. I’ll go on using Ursula as an example—see? Like this.”

“I don’t see . . . ” I began.

“Each figure is multiplied by the number at the head of its column,” said Scrivener patiently.

I looked again and understood. I nodded.

“You’d have to destroy all your workings out afterward,” he said, “but this way, you wouldn’t easily go wrong.”

“Let’s try something longer,” I said. “I want to understand this properly. I may not be able to decipher these letters—after all, there seem to be so many possible variations—but all this is interesting. I want to grasp the idea behind it properly. Let me see . . . ”

Scrivener chuckled. “Who shall we involve in this sorry conspiracy, you mean? The queen’s Sweet Robin, her Eyes, as I believe she variously calls him; shall we have him? He is not widely popular. Let us encipher the words
My Lord of Leicester.
We write it out—so—and then, under each letter, goes its first-stage number—still using the simple system where A equals 1.”

“Then,” he said, “you sort it all out into its four columns, so.”

We all studied this with interest and I picked up the abacus again. “Let me see.”

It was something of a struggle. My tongue came out between my teeth, as though I were a small child wrestling with a first attempt to
learn the alphabet. Finally, though, I sat back in triumph. After much muttering and clicking, because I was out of practice, I had arrived at 39, 100, 60, 90, 54, 16, 75, 36, 36, 20, 45, 18, 15, 76, 100, 30, 54.

I considered this with my head on one side. “There are some repeated figures. Thirty-six, for instance.”

“Yes, but those two thirty-sixes represent different letters. The first one is the F of OF and the second is the L of LORD.”

“But the R in LORD and at the end of LEICESTER both work out to 54.”

“It does happen,” Scrivener agreed. “But not too often.”

“My head’s spinning,” said Dale, making us all laugh.

And then it was Dale, who, leaning forward and pointing excitedly at the top sheet on the pile of coded documents, said: “But this one starts with just those figures! Look! 39, 100, 60, 90 . . . ”

 • • • 

We all stared at each other, excitement silently rising. “Can you test it, madam?” Brockley asked.

“The last letter in LEICESTER is R,” I said. “In stage one, that’s 18, and it comes under column one, so it’s multiplied by 3 and that’s”—I once more checked on the abacus—“54. Yes, there’s no doubt about that. So . . . so . . . ”

“The next figure to be deciphered has to be
divided
by 4,” said Scrivener. “Go on.”

“The next figure
is
4,” I said. “So that would be 1! The letter A, presumably, at least if this is just being done by the most straightforward system. Then what? The next figure is 95. I’ve got to divide that by . . . yes, I see, by 5. That’s . . . that’s . . . ” The abacus clicked furiously. “That’s 19,” I said triumphantly. “That’s S. Then comes 36, which has to be divided by 6—which
is
6. That’s F. Then . . . ”

A few minutes later, I had the sequence 1, 19, 6, 1, 18, 1, and 19 and had written out their alphabetic equivalents.

“As far as!”
Brockley almost shouted.