Philosophy of Science
Muhammad Sajeer Bukhari
Chapter 01
Induction and Naive Inductivism
Introduction
This chapter explores major developments in the philosophy of science, with a focus on induction and naive inductivism. The aim is to provide a foundational understanding of how science generates knowledge, which is crucial for grasping the scientific method.
The Power of Science
Science has made enormous progress. We seem to understand the world far better today than we did 300 years ago. We have incredible abilities to control and manipulate nature, and we've been able to make massive improvements to human lives as a result. This progress is almost entirely because of science. So, the question naturally arises: How exactly has science achieved this? What is it about science that allows it to exhibit this power? One way to put this is: What is the scientific method? This is the sort of thing we'll be looking at.
Nature of Confirmation
A key feature of science is its ability to generate theories and test them through observations. Certain observations support or confirm these theories. This chapter examines two proposed methods for relating observations to theory: the naive inductivist method and the hypothetico-deductive method. These methods align closely with common perceptions of the scientific method, making them a suitable starting point..
Types of Inference
Before exploring these methods, it is essential to understand different kinds of inference. First, there is deductive inference or deduction. In a deductive inference, the truth of the conclusion is guaranteed. If the premises are true, the conclusion must be true. The classic example is: "All men are mortal, Jesus is a man, therefore Jesus is mortal." If the premises of this argument are true, the conclusion must be true. It cannot fail to be true. There is no possible way that the premises could be true and the conclusion false. You can't even imagine a situation in which these premises are true and the conclusion is false.
Deduction vs. Induction
The key aspect of deduction is that it is non-ampliative, meaning the conclusion does not contain any additional information beyond what is already present in the premises. The conclusion highlights some of the information already asserted by the premises. Therefore, the truth of the premises ensures the truth of the conclusion because the conclusion simply reasserts what has been stated in the premises. Consequently, if the premises are true, the conclusion must also be true. Deduction is crucial in logic and mathematics but is less central to the sciences. Science seeks to uncover new information about the world, which is not achieved through deduction alone.
Induction
In contrast, Science relies on non-deductive inferences, commonly known as induction. Classic examples of induction include statements like: "All observed swans have been white, therefore all swans are white," or "All observed swans have been white, therefore the next swan observed will be white." In such cases, the premises do not guarantee the truth of the conclusion. For instance, observing that swan number one is white, swan number two is white, swan number three is white, and so forth, does not preclude the possibility that the next swan observed could be black. The claim that all swans are white extends beyond the observation that all observed swans are white.
Enumerative Induction
Throughout this series, the term induction will be used to encompass all forms of non-deductive inference. There are various types of induction. The previously mentioned example with the swans illustrates enumerative induction. In enumerative induction, reasoning proceeds from the premise that all observed F's are G's to the conclusion that all F's are G's. However, science also relies on a different form of reasoning known as inference to the best explanation.
Inference to the Best Explanation
In this form of inference, the claim is made that if X is the best explanation for some phenomenon, then X is true. For instance, consider the extinction of dinosaurs 65 million years ago. To understand why they went extinct, one might note the presence of high levels of iridium in crust layers dating back to that period. Asteroid impacts are known to cause both extinction events and elevated levels of iridium, as asteroids contain significantly more iridium than the Earth's crust. Thus, a coherent explanation for these seemingly unrelated facts emerges: 65 million years ago, an asteroid struck the Earth, leading to both the extinction of the dinosaurs and the high iridium levels in the crust. The best explanation of these facts is the occurrence of an asteroid impact 65 million years ago. Consequently, this is the explanation that should be accepted as true.
Abduction
Inference to the best explanation, sometimes referred to as abduction, is often distinguished from both deduction and induction. For the purposes of this discussion, however, the term induction will encompass all forms of non-deductive inference. The essential point is that scientific inferences are ampliative; they extend beyond the information contained in the premises. Science is not merely a compilation of observations and their reiteration in different forms. Instead, scientific practice involves making observations and using these observations to support broader conclusions—conclusions about unobserved phenomena or future observations. In this way, science transcends the immediate evidence.
Naive Inductivism
Keeping that in perspective, let's outline the process. How does the scientific method unfold? First, we'll delve into naive inductivism, commonly attributed to the philosopher Francis Bacon. However, Bacon's stance might have been more nuanced than commonly believed. Nevertheless, naive inductivism typically comprises three fundamental steps:
1. Observation and Documentation of Facts: Initially, the aim is to observe and document as many relevant facts as feasible regarding the phenomenon under scrutiny. Ideally, one would capture all pertinent data, although this is often impractical. The emphasis lies on maintaining neutrality, avoiding preconceived theories. Instead, one simply records sensory perceptions—what is seen, heard, felt, or reported by measuring instruments. Experiments play a pivotal role in this phase, facilitating the exploration of phenomena under novel conditions, yielding fresh insights.
2. Analysis and Categorization of Facts: Subsequently, the accumulated facts undergo analysis and coherent classification. Bacon advocated for organizing facts into structured tables. For instance, in the study of heat, one might compile tables contrasting hot and non-hot entities, gauging the varying degrees of heat across different entities. This process abstains from presupposing any theoretical framework, focusing solely on enumerating instances of property presence or absence.
3. Formulation of Generalizations: Finally, from the pool of observed and classified facts, overarching generalizations are derived. Through meticulous examination, one identifies recurring patterns and associations. For instance, in the study of heat, one might inquire into the additional properties accompanying heat presence or absence. This systematic approach enables the generation of generalized principles regarding heat, which in turn serve as the basis for formulating and testing predictions.
Example of Naive Inductivism
Let's consider a scenario where the focus is on studying human life expectancy and the factors influencing it. Initially, researchers observe humans extensively, strictly documenting various aspects related to their lives. Next, they undertake the task of categorizing the amassed data, potentially segmenting individuals into distinct groups based on their smoking habits—such as smokers and non-smokers—and assessing whether they develop lung cancer. Subsequently, employing inductive reasoning, researchers draw generalizations from their observations. Notably, they may discern that smokers exhibit elevated rates of lung cancer incidence, thereby prompting the formulation of a hypothesis positing a causal link between smoking and cancer development—a hypothesis that can subsequently be empirically tested and validated.
Critical Examination of Naive Inductivism
Despite its initial appeal, this method encounters numerous challenges, as illuminated by Carl Hempel in "Philosophy of Natural Science." The primary hurdle arises during the initial phase of assembling a comprehensive set of facts impartially, yet the sheer multitude of potential facts presents a daunting task. Discerning the relevance of each datum becomes a perplexing endeavor. The pursuit of neutrality in data collection may inadvertently steer researchers towards collecting inconsequential information, a concern underscored by Charles Darwin. Darwin emphasized that every observation must serve to either support or refute a particular viewpoint. The absence of a guiding hypothesis complicates the selection of pertinent data, rendering the process ambiguous.
Problems with Classification
Likewise, the impartial classification of data in the second stage poses challenges. The potential for data to be categorized in countless irrelevant manners is evident, such as by trivial attributes like the number of hairs on an individual's arm or their distance from London. Moreover, pertinent classifications frequently defy conventional expectations. For instance, the proximity to populations of rodents emerges as a relevant factor in comprehending the spread of the plague, highlighting the unforeseen connections uncovered through assumptions regarding the transmission of bacteria by fleas on rodents.
Untrustworthy Observations
Certain observations prove unreliable, exemplified by the optical illusion where a stick submerged in water appears bent. The impartial reporting and classification of observations become challenging, particularly with advanced instruments susceptible to diverse errors. Consider microscopes, which demand meticulous sample preparation and alignment; any contamination can introduce distortions in the resulting images. The determination of trustworthy observations often hinges on pre-existing theoretical frameworks, underscoring the interplay between observation and theoretical understanding in scientific inquiry.
Deriving Theories from Observations
A further challenge lies in the process of deriving theories from observations. Philosophers draw a distinction between observational and theoretical terms: the former refer to direct experiences, such as cars and mountains, while the latter denote entities like beta particles and neutrons, which elude direct observation. Crafting theories transcends mere generalizations from observations; it demands creativity and innovation, often manifesting as educated conjectures. Theoretical frameworks thus extend beyond empirical data, incorporating speculative elements essential for scientific progress.
Given these points, it becomes evident that naive inductivism is fundamentally flawed. Instead of beginning with observations and then forming hypotheses, scientific inquiry should commence with hypotheses to guide observations. Though this approach can introduce biases and errors, it remains the most viable option for modern science. The hypothetico-deductive method addresses these shortcomings and provides a stronger framework for scientific investigation, helping to resolve the issues with the naive inductivist approach.