Science is an ever-shifting miasma wherein the greatest minds recognized by history are exalted for their discovery and have their faults discarded. The term “biology” was coined long after the works of Aristotle were widely discredited, and he still retains the ambiguous title of “father of biology” (McRae 2010). This phenomenon of forgiveness is not limited to biology alone, as Einstein was incorrect in his postulation of a static universe and Pauling’s misconception of a triple helix (Olby 1974). This paper reviews the chronology of biological advancement, beginning Aristotle’s views on empiricism and classification through Popper’s ideas of falsification.
Prior to Aristotle, infrequent but significant progress on the scientific method occurred throughout history, including descriptions of both controlled experiments by Greek physicians (Algra 2003) and descriptions of Egyptian surgical practices (van Middendorp et al. 2010). These discoveries preceded Aristotle and his elements of deduction and sensory perception outlined in the Posterior Analytics (Pellegrin and Preus 1986).
According to the Lennox (2001), the First Principle outlined in Posterior Analytics is part of a larger syllogism that dictates in order for us to know something, it must be observable in some way. For instance, if water is wet, this idea must be demonstrable by observation.
Aristotle continued this tradition First Principle teleologically when he formed the Great Chain of Being (Govier 1997). All form must follow function and this function must be demonstrable (Depew 2008).
Why parts of Aristotle’s incorrect teachings like went so long unopposed is unclear. Gould (2002) and other prominent debate the veracity of claims of religious influence on scientific suppression during the Middle Ages. The printing press did allow for quicker dissemination and dismissal of Aristotle’s more floundering ideas (Eisenstein 1980), like the assertion that men had more teeth than women (Depew 2008).
Aristotle’s beliefs were modified by the Persian Avicenna and the Egyptian Alhazen, both clinical practitioners in the 11 th century. The former critiqued the First Principle of Aristotle, referring to deduction without experimental foundation of lesser utility ( Kelishadi and Hatami 2012). Alhazen was known for anatomical experiments of the eye, many of which stand true in modern optometry (Howard 1996). Prior to and many centuries after this physical and controlled testing, many scholars dealt with thought experiments (Kheirandish 2009).
The Western idea of combining Aristotle’s dictum with experimentation was popularized by Roger Bacon, a 13 th century Franciscan monk. Secreta Secretorum was highly influential on Bacon, which is an alleged work of Aristotle (Charles 1960). In addition to translating the Greek and Arabic texts of the aforementioned clinicians, Bacon developed the nascent theory of repeatable hypotheses during experimentation (Raizman-Kedar 2009).
Independently, the first null hypothesis was put into practice in 1279 to measure the weight of coins, but not widely used in the scientific method until the 19 th century (Curran-Everett 2009). The idea of hypotheses helped shape the scientific methods used in modern science in the Renaissance such as Robert Boyle’s fledging idea of falsification and subsequent critical refutation of Roger Bacon by the Enlightenment’s William Whewell (Agassi 2010).
In the preface to Jardine and Silverthorne’s 2010 analysis of Novum Organum , it is said that Francis Bacon was a 17 th century Irish-English polymath who was the founder of empiricism. By popularizing the use of the “If, And, Then” form of statements, he set the table for the dismissal of Aristotle’s deductive ideas. In conjunction with observations, this is known as induction. Hypotheses were still excluded from this new scientific paradigm (Fattori 2005). Francis Bacon also helped foster a few ideas still in use to this day, such as peer review (Spier 2002) and recognized the possible incidence of confirmation bias (Friedrich 1993).
Baxy (1997) reported single-blind testing in humans was not documented until the 18 th
century, with James Lind dividing sailors into groups of two to test the effect of acidic food on
scurvy. Lind did not know the mechanism of action, causality or that Vitamin C existed, as Vitamin C was not isolated until two centuries later (Carpenter 2012). Blinding of trials was proposed by the French Academy of Sciences in 1784 (Best et al. 2003).
Testing for causality in terms of disease was finally shifted against the spontaneous generation of disease by the postulates proposed by Robert Koch (Walker et al. 2006). Several scientists had run experiments or collected epidemiological data in attempting to disprove the miasma theory but never created a series of principles in the vein of Aristotle (Falkow 1988). A modified version of Koch’s postulates are still used today to identify causative pathological agents in microbial infection by genetic identification (Walker et al. 2006).
The first uncontrolled double-blinded placebo controlled trial published was detailed by Stoberg (2006). The trial was designed to test the efficacy of homeopathy using salt water, and gave crude results due to dropouts. The first robust double-blind trial, placebo-controlled trial occurred in the United States in the 1937 for tuberculosis vaccination, and remains the gold standard to this day (Best et al. 2003).
Meta-analysis was not used as a clinical tool until 1900 (O’Rourke 2007), when the first tables of typhoid vaccination efficacy were published without the statistical measurements discussed later popularized by Fisher. Fisher contributed study design in terms of randomization of trials and models used to test the null hypothesis in the form of ANOVA, Chi-squared and Fischer’s T-Test (Aldrich 1997).
A contemporary of these statisticians was Karl Popper, who became most known for his use of the null hypothesis, despite its prior description. In a recent review by Wilkinson (2012), it’s inferred he used deductive methods only, a premise contradicting Baconian empiricism and under debate (Lakatos 1974).
Several researchers have contributed to maximizing results via multiple hypotheses, a recent development in the 20 th century by Platt (Weisman 2008) and earlier by Chamberlain via
randomization (Dehue 1997) during the early stages of quantifying the null hypothesis.
Will biologists today still have the same turmoil as history has shown us? It’s impossible to know without pure speculation. Many of the studies are incorrect based on the mathematical models used to prove they are indeed correct, especially when sample sizes are low (Ioannidis 2005). This does not take into consideration of the drawer effect, where studies that succeed to null the hypothesis are published irregularly (Rosenthal 1979).
To borrow an axiom from Churchill: The scientific method is the best form of scientific method we have. While prior attempts to disprove biology status quo have been qualitative, it’s possible the new models of biology are combining other fields like computation and mathematics much like the scientific method borrowed from philosophy. These faults are part of a David Hume central thesis in paraphrase: what holds true today, may not hold true tomorrow (Morabia 2005).