Franziska Reinhard

Prototypical historical sciences are in the business of reconstructing the past based on traces. Experimental research into the past often seems precluded by the very nature of historical phenomena: they are not only unobservable, but unmanipulable. In this paper, I focus on a less prototypical historical science–origins-of-life research–to add to the debate over the role of experiments in the historical sciences. I identify and address two distinct challenges for experimental research into the deep past. The first, Turner’s challenge, concerns scientists’ inability to manipulate research targets in the deep past, while the second, Cleland’s challenge, questions whether experimentation can support inferences about such research targets. Turner’s challenge can be resolved by distinguishing between two experimental strategies–isolation and synthesis–that rely on different types of manipulability. While isolation depends on token-manipulability, synthesis requires type-manipulability. Turner’s challenge presupposes isolation as the default experimental strategy, but can be mitigated when recognizing the central role of synthetic experiments in studying the deep past. To address Cleland’s challenge, one needs to pay attention to the diachronic nature of experimentation. In origins-of-life research, this means focusing on the processes by which experimental results are turned into actual evidence for steps on the chemical path to life. Although individual experiments may not always be sufficient in yielding significant inferences about the past, the accumulation and integration of different results over time can support such inferences.

How, where, and why life first emerged are still open scientific questions. To answer them, origins-of-life (= OoL) research draws on a variety of approaches from disciplines such as chemistry, biology, geology, and even astronomy. The goal is to account for the processes that led from simple chemical components to the first forms of life on Earth more than 3.5 billion years ago. This puts OoL-research in a particularly challenging epistemic situation compared to other sciences. OoL-researchers cannot expect to find substantial traces of the formation of the first forms of life. In addition, their knowledge of environmental conditions on the Earth at the time of the OoL is uncertain. And yet, for all that, OoL-research is a very active field. This sets the topic for this thesis: I study, from a philosophy-of-science perspective, the strategies OoL-researchers use to counteract and overcome their unusually difficult epistemic starting point to generate knowledge about the OoL. The thesis draws on and contributes to philosophy-of-science literature in two main areas: (i) general philosophy of science, particularly discussions relating to experimentation and explanation; and (ii) the philosophy of the historical sciences, specifically debates about resources for evidential reasoning and the role of experimentation in studying phenomena in the deep past. In Chapter 2, I provide an overview of OoL-research and motivate a focus on OoL-research in the context of the philosophy of the historical sciences. In Chapter 3, I analyze a case study on identifying and interpreting traces of very early life. Based on that, I challenge a common preconception in existing characterizations of ‘trace’ in the philosophy of the historical sciences and introduce a distinction between what I call ‘direct’ and ‘circumstantial’ traces to better understand situations where evidence from traces is particularly scarce or ambiguous. Chapters 4 and 5 both engage with specific case studies from prebiotic chemistry, a subfield of OoL-research that is largely experimental and draws on strategies from synthetic organic chemistry. In Chapter 4, I characterize ‘synthesis’ as its own, as yet neglected experimental strategy, and analyze the epistemic value of synthetic experiments in OoL-research. My proposal calls into question the argument that experimentation as a method to study past phenomena is either impossible or inefficient. In Chapter 5, I put forward a distinction between different functions of possibilities in scientific explanations that I call ‘elucidating’ and ‘embedding’ and illustrate their relevance to OoL-research. Furthermore, I highlight how my distinction improves upon existing debates over the nature of ‘how-possibly’ explanations.

Existing characterizations of ‘trace’ in the philosophy of the historical sciences agree that traces need to be downstream of the long-past event under investigation. I argue that this misses an important type of trace used in historical reconstructions. Existing characterizations of traces focus on what I propose to call direct traces. What I call circumstantial traces (i) share a common cause with a past event and (ii) allow an inference to said event via an intermediate step. I illustrate the significance of checking the alignment between direct and circumstantial traces in historical reconstructions through a case study from (micro-)palaeontology.

Philosophers of science have variously tried to characterize how-possibly explanations (HPEs) and distinguish them from how-actually explanations (HAEs). I argue that existing contributions to this debate have failed to pay attention to the different, but complementary, functions possibilities play in scientific explanations. To bring these functions to the fore, I introduce a distinction between what I call elucidating and embedding HPEs. While elucidating HPEs specify and demonstrate possible processes for a given research target, embedding HPEs demonstrate how the research target fits into a space of suitably constrained possibilities. I specify both functions of HPE with reference to two case studies from origins-of-life research. I contrast my distinction with an alternative proposal by Wirling and Grüne-Yanoff (2024) to highlight that focusing on the functions, rather than types, of possibilities in explanation is better suited to account for key scientific practices.