Understanding the Complexities of Scientific Communication
Written on
Chapter 1: The Challenges of Science Communication
The current state of science communication (scicomm) is flawed, and this extends beyond the interactions between scientists and the general public. Academic papers often contain terminology that reflects a fundamental misunderstanding of the scientific process.
When examining the language used in science communication, it becomes clear that terms like confirmation and verification are misleading. Science, in essence, does not confirm anything; it primarily identifies when we are mistaken.
Section 1.1: A Philosophical Perspective on Science
To grasp the underlying issues with scicomm, it’s essential to touch on some foundational concepts in the philosophy of science. Historically, there was significant debate regarding whether scientific inquiries could indeed validate our hypotheses and predict future outcomes—a dilemma known as the problem of induction.
The prevailing resolution to this dilemma characterizes science as a process of falsification. In this model, we propose a theory, make predictions based on it, and subject those predictions to empirical tests. If our theory fails to align with observed results, it is deemed inconsistent with existing scientific knowledge.
Subsection 1.1.1: Hypothesis Testing in Science
The primary methodologies employed to evaluate the consistency of a theory against observed data include hypothesis testing through p-values and Bayesian inference. However, even this approach can lead to provisional falsification, as future data might ultimately validate a previously dismissed theory.
Section 1.2: Defining Robust Theories
While we can't assert that a theory is true or confirmed, the term "robust" serves as a more suitable descriptor. A theory's robustness correlates with its reliability, especially when alternative explanations have been thoroughly examined.
A robust theory must align with existing observations. However, if the data is sparse, labeling a theory as robust is unwarranted. A substantial amount of data is necessary, coupled with the theory’s capacity to generate diverse predictions. Moreover, the presence of competing theories that also fit the data reduces the robustness of any single theory.
Chapter 2: The Significance of Robust Theories
The robustness of a theory increases if it underpins numerous other theories. For instance, evolutionary theory is foundational for fields like psychology and medicine. Should the theory of common descent be disproven, it would necessitate a reevaluation of countless scientific frameworks.
Video Description: In this insightful discussion, Daniel Schmachtenberger delves into the concept of "Bend not Break," exploring the dynamics of maximum power and hyper agents in the context of scientific understanding.
Summary of Key Terminology
To recap the main points discussed:
- A theory is inconsistent with scientific observations if the observations appear unlikely, assuming the theory holds true.
- A theory is consistent with observations if they are probable under the assumption that the theory is accurate.
- A robust theory is one that is consistent with a significant body of observations, has undergone extensive testing, serves as a foundation for other theories, and faces minimal competition.
These definitions are not yet widely recognized, but I advocate for their adoption among science communicators and researchers. Until we find a solution to the problem of induction, we must accept that science operates as a system of falsification.
Further Reading
I have explored mathematical approaches to address the problem of induction, though the results are questionable and introduce other complexities.