In 2022, the Journal of Non-Newtonian Biomechanics published my paper “Volumetric Fluidity in Felids: Empirical Confirmation of the Liquid Cat Hypothesis.” It has since become the most-cited thing I have ever written. I am asked about it at every conference I attend. A journalist described it as “delightfully silly.” My postdoc supervisor — who spent fifteen years trying to get his own most-cited paper past four hundred citations — sent me a congratulations note that I am fairly confident was not entirely sincere.
I have complicated feelings about this, which I will now express at length.
What the Paper Actually Did
The premise of the Liquid Cat Hypothesis — that domestic cats conform volumetrically to any container — is, of course, a popular observation. It has been a meme. It has been a screen-saver. It is not, or was not, a quantified physical claim.
What the paper did was quantify it. Forty-seven container geometries. Twelve feline subjects. Skeletal compliance measurements. Musculoskeletal deformation rates. Post-transfer volume recovery. The container-filling efficiency of 98.7 ± 0.4%. The designation of the resulting class of matter as “bio-fluidic solids” — a term I coined in the third draft and which a referee asked me to define more precisely and which I defined more precisely and which is now in the literature and I cannot take it back.
The methodology was, and I say this without false modesty, excellent. The measurement protocol took eight months to develop. The statistical treatment was reviewed by a biostatistician who said it was “unusually careful for a paper with this title,” which I chose to take as a compliment.
Why It Is the Most-Cited
I have thought about this more than I should. The citation count is high for two reasons, and they are not the same reason.
The first is that the paper is, genuinely, good. It establishes a new class of biomechanical behaviour with a clean framework and reproducible results. Several subsequent papers on soft-body locomotion, postural mechanics, and bio-inspired robotics have cited it for the β_f parameter, which is the container-filling efficiency metric. These citations are what I would call correct citations. The work is being built upon.
The second reason is that the paper is about cats fitting into boxes. This makes it easy to describe at dinner parties, easy to share, and easy to remember. A disproportionate number of its citations are in introductions — the “delightful example” citation, which proves you have read widely and have a sense of humour, and then the paper moves on. These citations are not wrong, exactly, but they are not engaging with the science in any meaningful sense. They are engagement with the idea of the science, which is a different thing.
I am pleased by the first category of citation and ambivalent about the second, which I recognise is an ungracious position to hold about anything that helps my h-index.
What I Wish People Cited Instead
The 2021 paper on aeroacoustic signatures in barn owl feather microstructures, co-authored with M. D. Airquill and P. N. Silentswoop. It is technically superior, more difficult to have done, and cited by approximately one-third as many people. This is because “barn owl feather microstructures” is harder to explain at a dinner party than “cats are liquid.”
I have made peace with this. I have not entirely made peace with this.
The Owl–Cat Asymmetry
One finding from the Liquid Cat paper that attracted very little attention was the supplementary data on owl nesting boxes. In trials where the same feline subjects were introduced to standard Tyto alba nesting box geometries, the container-filling efficiency dropped to 94.2 ± 1.1% — still extraordinary by any measure, but significantly lower than the 98.7% achieved in purpose-built containers.
I proposed, in the discussion, that this might reflect active postural resistance in response to the olfactory cues associated with owl occupancy. Three people cited this. One of them was me, in a later paper. The other two are researchers I have met and like, and I am grateful to them.
The finding raises a question I have not fully answered: whether volumetric compliance in bio-fluidic solids is a purely mechanical property or whether it is modulated by behavioural context. I think it is the latter. I think the cat knows it is in an owl box and has opinions about this. I cannot prove the cat has opinions. But the data are consistent with a cat that has opinions.
The paper is open access. The cats are not available for follow-up questions.