How-to

How to Decrypt a Caesar Cipher by Hand

5 min read

A step-by-step guide to breaking a Caesar cipher with pen and paper by trying every shift, using letter frequencies, and testing likely words.

What is a Caesar cipher?

The Caesar cipher shifts every plaintext letter by the same number of positions in a chosen alphabet. With a shift of 3 in the 26-letter English alphabet, A becomes D, B becomes E, and X wraps around to A.

Because one fixed substitution is used throughout the message, Caesar is a monoalphabetic substitution cipher. The shift is the key. An alphabet of N symbols has N − 1 non-trivial shifts, so English has 25 candidates. That small keyspace makes exhaustive manual testing practical, although very short messages can have more than one plausible reading.

Before you start

First identify the alphabet used by the message. For this English example, write 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 on a strip of paper. A cipher made with another alphabet can have a different number and order of symbols.

To decrypt, move each ciphertext letter backward by the candidate key. A shift of 3 backward in a 26-letter alphabet is equivalent to 23 forward. Spaces, digits, and punctuation are normally preserved, but confirm that convention when the source is known.

Method 1 — Try every shift (brute force)

For an alphabet of N symbols, test shifts 1 through N − 1. Take the English ciphertext WKLV LV D WHVW:

  • Shift 1: VJKU KU C VGUV.
  • Shift 2: UIJT JT B UFTU.
  • Shift 3: THIS IS A TEST.

Shift 3 produces a clear sentence, so it is the strongest candidate. With a very short ciphertext, keep more than one plausible result until context confirms the answer. The Caesar brute-force tool lists every shift at once for checking your manual work.

Method 2 — Frequency analysis

Longer texts let you prioritize likely keys. In English, E, T, A, and O are common, but the exact order varies by sample. Count the ciphertext letters and compare several high-frequency candidates rather than trusting a single match.

If H is frequent and you hypothesize that it represents E, the distance is 3, so test key 3 on the whole message. Accept it only if words, grammar, and multiple letter patterns agree. The frequency analysis tool can produce the counts; frequency is a clue, not proof.

Method 3 — Use a crib (a likely word)

A likely word—a greeting, a name, or a term suggested by context—is called a crib. Align it with a ciphertext fragment and calculate the backward shift for every letter.

If you expect THE and see WKH, then W→T, K→H, and H→E all require a shift of 3. That consistency makes key 3 worth testing on the complete message. A crib is convincing only when the rest of the plaintext also becomes coherent.

Common mistakes and checks

Check these common failure points:

  • Wrong alphabet. The number and order of letters determine every shift.
  • Wrong direction or wraparound. In English, B shifted back by 3 is Y.
  • Premature certainty. A short result can look meaningful by chance; verify the entire message.
  • ROT13 scope. ROT13 is the self-inverse shift-13 case for the 26-letter Latin alphabet.

If no shift produces coherent text, reconsider the alphabet, the language, and the assumption that the message uses a Caesar cipher. It may instead use a different substitution or a transposition.

Frequently asked questions

An alphabet of N symbols has N − 1 non-trivial Caesar keys. The 26-letter English alphabet therefore has 25. Shift 0, or any full rotation, leaves the text unchanged.

Usually. You can test the complete keyspace, then rank the results using language, context, frequency patterns, and cribs. Extremely short or context-free messages may remain ambiguous even after every shift is listed.

ROT13 is a Caesar cipher with shift 13 over the 26-letter Latin alphabet. Because 13 is half of 26, applying it twice returns the original text. It can hide casual spoilers but provides no cryptographic security.

Short samples fluctuate: their most frequent letter may not match the language average. More ciphertext provides more stable counts, although genre, names, and repeated words can still skew the profile.

See also