If All Knowledge Is Provisional, When Can We Have Confidence in What We Claim to Know?

“The actual state of our knowledge is always provisional and… there must be, beyond what is actually known, immense new regions to discover.” Nobel laureate Louis de Broglie’s apt observation reveals the transient, or provisional, nature of knowledge. Provisional knowledge means it serves its function for a window of time, before it is revised and replaced. Paradigm shifts or improved evidence can bring new interpretations to even the most “objective” of facts. This “provisional” nature questions our confidence in a piece of knowledge. Confidence in knowing is a position where an individual has a strong belief that something is indeed true. Confidence may stem from personal observations or shared knowledge. In this essay, two areas of knowledge, namely Natural Science and History, will be investigated on the degree of confidence given their provisional nature.

Natural Science encompasses many fields united by the scientific method, which is the process used to generate scientific knowledge. It involves the identification of a question and hence a hypothesis. Experiments involving controlled and systematic observation are subsequently used to amass empirical data. The analysis of empirical data yields a scientific theory, which is a well-substantiated explanation that can repeatedly be tested and verified with further replication. Empirical evidence is the centrepiece of the scientific method, meaning that it is a posteriori knowledge. Moreover, a key element of theories is that they have to be falsifiable. This implies that theories can be revised in light of new data; At no stage is absolute truth guaranteed.

The methodology of science is what truly solidifies our confidence in scientific knowledge. Indeed, the scientific method, with its empirical emphasis and objective testing of hypotheses, has allowed humanity to surpass superstitions that were unreliable. Investigations are expected to keep personal biases, false beliefs and emotion at a minimum. If an isolated research paper was to meet these criteria of objectivity, it already gives grounds for trust in its knowledge claims. In addition, replication also reinforces confidence. It allows theories to be tested and validated by other researchers instead of simply being conjectures that could never be verified or falsified. Scientists submit their findings to other scientists for analysis, critique and even experimental replication. This process is known as peer review. Over the long-term, a consensus about scientific conclusions emerges from the peer-review process. More importantly, our confidence in natural sciences can come from its pragmatic nature. It has been established that "truths" in natural science are never absolute. Despite this, science is still used in everyday life. In 1662, Robert Boyle established an inverse relationship between pressure and volume in his second edition of The Spring and Weight of the Air. This relationship is known as Boyle’s law. These observations have been replicated worldwide. Thus far, all forms of replication have verified this theory. Given that this empirical relation has stood the test of time, it has been heavily applied in the development of technologies in engineering. Only what has stood the test of time as useful becomes incorporated into the larger scientific consensus. Boyle's law faces the perpetual risk of being falsified. However, no instance of replication had yielded incongruous evidence that proves that it is false. The perseverance of validity achieves one of the greatest forms of scientific confidence. Therefore, scientific confidence is reinforced if knowledge can be harnessed to advance society.

However, science is a human endeavour, and experimental design and analyses are products of human thought processes. It is not infallible. Hence, “bad science”, or science with flawed and biased methodology, is bound to exist. Although scientists should strive for impartiality, it is challenging to completely avoid the subconscious influences of cognitive bias that undermine scientific validity. A form of this is confirmation bias, or the propensity to seek or interpret evidence in ways that are partial to existing expectations. Scientists may choose to disregard or overemphasise certain data due to personal beliefs or perceptions. For example, before Einstein established his theory of relativity, the scientific consensus was that light travelled through a vacuum via a theoretical medium, the “Aether”. A lack of evidence gave way for Einstein to propose a much stronger theory. However, conservatives in the community were against this revolutionary theory and began to scour for rebuttals. The rebuttal was found in an experiment conducted by E.W. Silvertooth, producing results that seemed to “prove Einstein wrong”. Nevertheless, Doug Marett repeated the experiment in 2012, finding that poor experimental design caused “significant data” to be due to temperature fluctuations. Silvertooth’s incorrect analysis illustrates how evidence was used to fit hypotheses instead of the reverse. Presently, the scientific field also involves a competitive culture that rewards piquant findings and undervalues replicating controversial experiments. The overemphasis on high-impact research incentivises researchers to compromise experimental rigour to fit results. Overall, these elements compromise the integrity of the scientific method and reduce our confidence in scientific knowledge.

History is an area of knowledge that offers a different perspective of confidence in provisional knowledge. History is the realm of knowledge of the past. Therefore, historians can only construct plausible explanations for evidence that the past has left behind. This domain of evidence includes first-hand accounts, archives and physical evidence. It is important to note that historians, though presented with the same evidence, may yet provide different interpretations. Therefore, history is provisional, subject to changes in the evidence or the interpretation.

Nevertheless, the transience of historical interpretations cannot disregard the rigour of historical investigations. Our confidence in historical knowledge is strengthened by the reliability of investigations. Critical analysis and multiperspectivity, which involves incorporating a diverse collection of viewpoints, are the bedrock of reliability. As the validity of a one-sided narrative is easily put under question, employing multiple perspectives yields a more impartial picture. For example, in Res Gestae, former Roman emperor Augustus states that “on my own initiative and at my own expense, I raised an army by means of which I restored liberty to the republic, which had been oppressed by the tyranny of a faction.” However, the reality was that Augustus had no legal right to command troops, and the tyrannical “faction” was actually led by the head of state Mark Antony. This was an armed insurrection against Rome, of which Augustus emerged victorious. Thankfully, an overwhelming amount of evidence produced by other Romans - Cicero's contemporary speeches, Suetonius, and Cassius Dio etc - revealed Augustus’ ulterior motive to protect his reputation. From Augustus’ example, it is evident that multiperspectivity constructs an accurate and coherent event by corroborating claims from different sources. Moreover, this illustrates how our confidence in different sources will vary. Multiperspectivity does not mean that it falls into the trap of relativism, where all sources have to be treated equally, eliminating the notion of historical truth. Instead, some sources are more reliable than others. Historians can determine this by questioning, the agenda, audience or the aspects that a source emphasises. Ultimately, a robust investigative methodology strengthens our confidence in a historical assertion.

Nonetheless, historical knowledge is constrained by human failings, weakening our confidence in a claim. Personal recounts may be intentionally or unintentionally distorted by memory or the passages of time. The personalisation of events may ultimately diminish or over-simplify the event. Additionally, an individual’s perspectives are moulded by their social and political inclinations. Moreover, it is often said that “History is written by the victors''. History has a tendency to exclude minorities in evidence collection. For example, the 1963 internal security dragnet Operation Coldstore, where 113 members of the left-wing Barisan Socialis were detained. The narrative that the government upheld was that they were involved in a violent Communist conspiracy to subvert authorities. However, historian Dr Thum Ping Tjin asserted that authorities had disseminated falsehoods and Coldstore was rooted in more political rather than security reasons. This sparked a debate that has brought doubt into an event that previously possessed a high degree of confidence. As with the example, history can be abused by those in power. The source material for the narrative of the “losers” is often lacking in quantity and quality compared to the winning side. Moreover, an event can be painted in a certain way to fit a political agenda. Therefore, the degree of confidence is influenced by a complex set of factors of memory, emotion or an agenda.

In conclusion, a probe into the areas of knowledge discussed reveals that we can never attain full confidence. Confidence is strengthened by a robust methodological foundation but yet shaken by procedural or evidential deficiencies. Strong investigative procedures and the incorporation of multiple perspectives may allow a piece of knowledge to be proven beyond a reasonable doubt. In addition, we cannot disregard the power of knowledge based on its utility in the real world. Knowledge pragmatism can function as a reference point when determining our confidence in the subject matter. We can assess “confidence” based on its utility within the present day cultural, political and economic circumstances. If what we think we know does not yield the expected results, further inquiry and discourse is in order. While it does not guarantee that inquiry will reach a definite conclusion, emphasis on utility strongly suggests that our inability to attain full confidence will be but a minor hindrance in our pursuit of knowledge. Humanity should strive for absolute truth, though the reality is that our confidence would ultimately lie at a satisfactory level. Embracing this notion shifts our focus onto persistent discovery, to continually push ourselves beyond the established limits.