Engineers working on the Harvard Mark II computer in 1947 found a moth jammed inside the machine’s relay hardware. They taped the dead insect into the operations log book and scrawled a now-famous annotation beside it: “first actual case of bug being found.” That single page, preserved for decades at the Smithsonian National Museum of American History, became the origin story most programmers and students encounter when they first hear the word “bug” applied to a computer fault.
How a dead moth shaped the language of software failure
The 1947 moth incident did not invent the word “bug” for a mechanical defect. Thomas Edison and other inventors used the term decades earlier to describe unexplained faults in electrical equipment. What the Harvard Mark II log entry did was anchor the metaphor to a literal, physical specimen and place it inside the emerging field of electronic computing. That combination of timing and tangibility gave the story staying power that no earlier usage could match.
The Smithsonian has reinforced that staying power through repeated public display. The museum mounted an exhibition that presented the moth found at the Harvard Computation Laboratory in 1947 as the canonical “first, actual bug” discovered in a computing device. By placing the artifact in a national museum setting and framing it with that specific language, the Smithsonian gave the anecdote institutional weight. Computer science textbooks, coding bootcamp slide decks, and conference keynotes have echoed the story ever since, often treating it as the definitive origin of the term rather than one colorful episode in a longer linguistic history.
The practical result is that the word “bug” now carries an almost universal meaning in technology. When a banking app crashes or a medical device misfires, engineers, regulators, and ordinary users all reach for the same vocabulary. That shared language traces a direct line back to a single taped moth on a single log book page, amplified by decades of museum curation and retelling.
What the Smithsonian’s log book page actually shows
The physical evidence is a single page from an operations log book used at the Harvard Computation Laboratory. Cataloged as object NMAH 334663, the page holds a small moth affixed with tape and the handwritten note “first actual case of bug being found,” according to the Smithsonian’s description of the log book page. The entry documents a real hardware interruption: the moth had lodged in one of the Mark II’s electromechanical relays, preventing the contacts from closing properly.
Curatorial notes from the Smithsonian add an important detail about attribution. Grace Murray Hopper, the Navy officer and computer scientist who spent years working on Harvard’s early machines, is often credited with discovering the moth and coining the phrase. The institutional record tells a different story. The museum’s own catalog entry states that the log book is probably not Hopper’s. Engineers on the Mark II team found the moth and made the entry; Hopper later popularized the anecdote in lectures and interviews, which cemented her association with the event in popular memory.
Archival finding aids for the Grace Murray Hopper Collection, which covers the period 1944 to 1965, contain primary documents and photographs related to Hopper’s Mark II work. Those records help researchers trace what Hopper actually contributed to the project versus what was attributed to her after the fact. The log book page itself sits in the museum’s general holdings, not within Hopper’s personal papers, reinforcing the distinction between the team’s discovery and one individual’s later storytelling.
Gaps in the moth story that still lack answers
Several basic facts about the 1947 incident remain unresolved. The surviving log book page and the Smithsonian’s catalog entries do not record the exact calendar date or the shift time when the moth was found. Accounts in secondary literature have offered dates ranging across the late summer and early fall of 1947, but no primary document in the museum’s holdings pins down the moment with certainty.
Equally unclear is who, specifically, wrote the annotation. The Mark II team included multiple engineers and technicians. No statement from any other team member has surfaced to confirm authorship of the note or to describe how the story first spread beyond the lab. The log book entry is, in effect, an unsigned communal record rather than a signed personal account.
The Smithsonian’s decision to acquire and repeatedly exhibit the artifact also lacks a documented paper trail in publicly available institutional records. Internal acquisition memos or curatorial justification files have not appeared in the finding aids reviewed for this reporting. Without those records, the reasoning behind the museum’s choice to treat this particular page as a signature object of computing history is a matter of inference rather than documented institutional policy.
For readers who teach, study, or simply repeat the moth story, the practical takeaway is straightforward. The physical artifact is real, the handwritten note is genuine, and the broader narrative has been shaped by decades of retelling. What remains uncertain are the finer-grained details historians usually want: who wrote the words, exactly when they wrote them, and how the joke traveled from a lab log to a global metaphor.
Why the myth persists despite the gaps
Stories like the Mark II moth endure because they compress a complex technical idea into a vivid, human-scale image. Debugging an electromechanical relay or tracing a software fault through thousands of lines of code is abstract and tedious. A moth stuck between contacts is concrete, even funny. It gives students and non-specialists a way to visualize failure in a machine that would otherwise be opaque.
The anecdote also fits a familiar pattern in the history of technology: a single, almost accidental moment becomes a stand-in for slower, more distributed change. The language of “bugs” and “debugging” evolved over decades of engineering practice, but the 1947 incident offers a neat origin point that can be told in a few sentences. That neatness makes it attractive in classrooms, documentaries, and museum labels, even when historians know the reality is messier.
Institutional endorsement magnifies that effect. Once a national museum labels an object as the “first, actual bug,” the phrase takes on an authority that is hard to dislodge. Visitors see the moth, read the caption, and carry the story outward. Each subsequent repetition flattens nuance: the distinction between first documented bug in a computer and first use of “bug” in engineering, or between a team’s in-joke and a formal technical term, tends to disappear.
Teaching with the moth, not just about it
For educators and communicators, the challenge is not to discard the moth story but to use it more carefully. Presenting the log book page as one colorful episode in a longer history of engineering language allows students to appreciate both the power and the limits of origin myths. It becomes a starting point for discussing how terms migrate across fields, how institutions curate narratives, and how individual figures like Hopper can become attached to events they did not directly author.
In practical terms, that might mean pairing the anecdote with earlier uses of “bug” in electrical engineering, or asking students to examine how museum catalog records differ from popular accounts. It can also open conversations about credit and collaboration in technical work: who gets remembered, who gets written out, and how archival decisions shape those memories.
The moth taped into the Harvard Mark II log book will likely remain a star exhibit in computing history for years to come. Its appeal is understandable: a fragile insect, a wry caption, and a direct link to the birth of modern computing make for an irresistible combination. A more careful reading of the available records does not diminish that appeal; it simply situates the insect where it belongs, as a tangible artifact in a much larger and more complicated story about how we name, record, and remember failure in our machines.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
