On the Limits of Experimental Knowledge

To demarcate the limits of experimental knowledge we probe the limits of what might be called an experiment. By appeal to examples of scientific practice from astrophysics and analogue gravity, we demonstrate that the reliability of knowledge regarding certain phenomena gained from an experiment is not circumscribed by the manipulability or accessibility of the target phenomena. Rather, the limits of experimental knowledge are set by the extent to which strategies for what we call `inductive triangulation’ are available: that is, the validation of the mode of inductive reasoning involved in the source-target inference via appeal to one or more distinct and independent modes of inductive reasoning. When such strategies are able to partially mitigate reasonable doubt, we can take a theory regarding the phenomena to be well supported by experiment. When such strategies are able to fully mitigate reasonable doubt, we can take a theory regarding the phenomena to be established by experiment. There are good reasons to expect the next generation of analogue experiments to provide genuine knowledge of unmanipulable and inaccessible phenomena such that the relevant theories can be understood as well supported.

💡 Research Summary

The paper argues that the limits of experimental knowledge are not determined by whether the target phenomenon can be directly manipulated or observed, but by the availability of “inductive triangulation” – the use of multiple, independent modes of inductive reasoning to mitigate reasonable doubt. The authors first distinguish internal validity (justifying claims about the source system that is directly manipulated) from external validity (justifying claims about a target system that is not directly manipulated). Conventional empiricism holds that external validity requires the target to be manipulable or at least observable; the authors reject this, claiming that external validity can be secured whenever independent inductive strategies can cross‑validate the source‑target inference.

Inductive triangulation is defined as the combination of distinct uniformity principles—temporal, spatial, inter‑type, etc.—each providing an independent line of support for the same conclusion. When these lines of support partially reduce reasonable doubt, a theory is “well‑supported”; when they fully eliminate reasonable doubt, the theory is “established” by experiment.

Two case studies illustrate the framework. The first concerns stellar nucleosynthesis. The nuclear processes in stellar cores are inaccessible, yet three independent strands of evidence—astronomical observations (spectra, supernova remnants), terrestrial nuclear experiments, and theoretical modeling—provide mutually reinforcing support based on different uniformity principles. This triangulation grants the nucleosynthesis theory full external validity despite the target’s inaccessibility.

The second case examines analogue gravity experiments aimed at confirming Hawking radiation. Laboratory analogues (e.g., acoustic horizons in fluids, optical analogues, Bose‑Einstein condensates) display Hawking‑like emission. The authors show that three independent justifications—mathematical analogy between general relativity and the analogue system, replication across different analogue platforms, and consistency with astrophysical predictions—constitute inductive triangulation. They further distinguish “rule circularity” (non‑vicious reliance on a reasoning rule to justify that rule) from “premise circularity” (vicious fallacy). The analogue approach relies on rule‑circular but non‑vicious reasoning, which is acceptable when bolstered by independent uniformity arguments.

From these analyses the authors draw three core claims: (1) Theories about unmanipulable, inaccessible phenomena can be experimentally established if sufficient inductive triangulation is available. (2) Experiments on manipulable source systems can yield inductive inferences about target systems of a different type, provided an inter‑type uniformity principle is justified. (3) Future generations of analogue experiments, by employing richer source systems and multiple uniformity principles, can move many currently “partially mitigated” theories into the “fully established” category.

In conclusion, the epistemic boundary of experimental science is set not by physical access but by the methodological resources for cross‑validating inductive inferences. When diverse, independent inductive routes converge, reasonable doubt is reduced to the point where a theory can be said to be established by experiment, even if the phenomenon itself remains beyond direct manipulation or observation. The paper predicts that forthcoming analogue‑gravity experiments will provide genuine knowledge of phenomena such as Hawking radiation, thereby expanding the scope of experimentally grounded science.