Do Old Books Smell Like Chocolate? Research on Recording Scents

When people smelled a sample made from an old book without being told what it was, the most common word they offered was “chocolate,” followed by “coffee.” In contrast, in a small evaluation involving an actual visit to a historic library, everyone described the smell as “woody.”

The 2017 paper reporting this surprising difference was not simply a study of what old books smell like. It treated easily lost smells as cultural heritage and proposed a framework for recording and preserving their value, chemical components, and the ways people experience them.

Before asking what it smells like, ask why it is worth preserving

The authors proposed four stages for working with historic odors: assessing cultural significance, chemical analysis, sensory evaluation, and documentation and preservation. Rather than collecting every smell associated with old things, the approach begins by considering what the smell means to a place, an activity, or people’s memories.

The smell of old paper and a library was chosen as a case study. Visitors to the library at St Paul’s Cathedral in London had written favorably about its smell, suggesting a connection between the value of the place and the sensations experienced there.

Odor compounds are also associated with paper degradation. The study thus sought to connect the recording of cultural meaning with an understanding of the physical condition of objects.

Chemical analysis of an old book and library air

The book sampled was Panait Istrati’s novel Les Chardons du Baragan, published in Paris in 1928 and obtained from a secondhand bookshop in London. It was left in a bag at room temperature for two weeks, and volatile compounds accumulating around it were collected on a thin sampling fiber for analysis. Airborne compounds were also sampled in the library.

The method combined solid-phase microextraction with gas chromatography–mass spectrometry. Examining the air around the book allows compounds to be recorded without damaging the object. The old book and the library shared many compounds, which the paper brought together in a single table.

The table includes acetic acid, hexanal, furfural, benzaldehyde, and other compounds. Existing odor databases associate furfural with descriptions such as bread, almond, and sweet, and benzaldehyde with almond and burnt sugar.

However, this matched detected compounds to known odor descriptions. The researchers did not perform calibration to quantify the concentration of each compound or test whether recombining the compounds recreated the book’s smell. How much each molecule contributes to the actual impression requires separate investigation.

Without being told it was a book, people thought of chocolate and coffee

At Birmingham Museum and Art Gallery, the researchers collected 120 descriptors from 79 visitors for the sample derived from an old book. The containers carried only letter labels, with no visual or verbal clues that the contents came from a book.

The sample smelled here was not simply the air released when opening an old book. Old paper was heated at 100°C for 2 hours and then extracted with a solvent. The extract was applied to gauze, and the solvent was allowed to evaporate before presentation. This preparation matters when considering how the sample relates to the everyday smell of books.

“Chocolate” appeared 21 times, or 27 when related descriptions such as cocoa were included. Coffee followed with 10 occurrences, old with eight, and wood and burnt with four each. These are counts of descriptors, not a result showing that everyone answered chocolate.

The same exhibition presented eight odors in total, including chocolate and coffee. The authors acknowledged that this experience could have influenced the choice of words. At the same time, some volatile compounds produced as paper degrades are also found in the aromas of coffee and cocoa, so chemical similarities offer another possible explanation. This survey cannot separate the influence of shared compounds from that of the exhibition context.

In the actual library, woody, smoky, and earthy impressions stood out

In a separate evaluation, seven untrained people entered the library at St Paul’s Cathedral and described its smell using 21 predefined terms and a space for free responses.

All seven selected “woody,” followed by “smoky” and “earthy.” Other descriptions included sweet, musty, and almond-like. Mean odor intensity fell between “strong” and “very strong” on the study’s scale, and most evaluators rated the smell as pleasant.

In this setting, bookshelves and wooden furniture were visible, and participants were supplied with words to choose from. The museum sample and the library differed in the odor being evaluated, presentation, question format, and participants. The two findings therefore cannot be treated as a controlled experiment showing that seeing the surroundings alone changed a chocolate smell into a woody one.

A Historic Book Odour Wheel connecting compounds and words

Combining these descriptions with chemical analysis, the researchers proposed a recording tool called the Historic Book Odour Wheel. It links broad odor categories, more specific descriptions within them, and potentially related compounds.

For example, terms such as old room, damp, and musty can be organized into a broader category and linked to analytical findings and existing odor information. The aim is to make it easier for both specialists and members of the public to put perceived smells into words and share them with conservators and chemical-analysis experts.

The wheel is not a finished recipe for recreating a smell or a test that diagnoses paper degradation by smell alone. It is a starting point for documentation that can be updated as new measurements and descriptions become available. Recording who smelled something, where, and with what contextual clues also makes it possible to address aspects of the experience that chemical components alone cannot preserve.

Summary

An extract derived from an old book and presented without revealing its identity elicited many descriptions of chocolate and coffee. In an actual historic library, woody, smoky, and earthy impressions stood out. Chemical analysis also recorded volatile compounds shared by the old book and the library.

The study’s central proposal is that preserving a smell requires more than a list of compounds: its cultural meaning and the ways people experienced it should also be recorded. The Historic Book Odour Wheel was a step toward turning a fleeting sensation into material that can be discussed and investigated again.

The research was supported by the UK EPSRC Centre for Doctoral Training in Science and Engineering in Arts, Heritage and Archaeology and by the National Trust. The authors declared no competing interests.

Reference

Bembibre C, Strlič M. Smell of heritage: a framework for the identification, analysis and archival of historic odours. Heritage Science. 2017;5:2. doi:10.1186/s40494-016-0114-1