How doctors think is an important issue for many healthcare professionals, especially in terms of cognitive mistakes and errors, and is a title of two recent books (Montgomery, 2006; Groopman, 2007). Biomedical practitioners generally subscribe to an objective way of thinking or reasoning that takes science as its example of how best to obtain and substantiate knowledge. Such knowledge is impersonal and has been described as "the view from nowhere" (Nagel, 1989).' In other words, this knowledge is applicable and valid for all times and places, regardless of one's particular values or biases or cultural context. Objective thinking brackets the emotions and intuitions, which proponents claim distort our knowledge of the world. "Intuitive thinking, brainstorming, creative option generation, and open-ended questions," for the biomedical practitioner according to Davis-Floyd and St. John, "are usually taboo" (1998, p. 33).
Humanistic or humane practitioners, although recognizing the significance and value of objective knowledge for medical practice, subscribe to a subjective way of thinking and reasoning that includes the intuitions, values, and virtues of the knower. Importantly, this type of thinking, especially in medicine, is based on the patient's narrative of the illness experience, as well on the physician's personal narrative of what it means to be a healer. In this chapter, I discuss objective thinking and reasoning in terms of the debate over the empirical and rational justification of knowledge and with respect to the logical nature of knowing. The subjective way of thinking and reasoning is discussed in terms of intuitions, values, and virtues, as well as narrative. In this way of thinking, humane practitioners address the quality-of-care crisis.
6.1 Objective Thinking
Objective, or scientific or impersonal, ways of thinking or reasoning are concerned with generating knowledge that is universally true about the world. This knowledge is taken to be factual and the facts that make it up are thought to be value-free, i.e. facts that are not distorted by predetermined conceptions of how the world is.' "Facts," according to Cassell, "can be verified-empirically demonstrated; everything that is not a fact is unavoidably doubtful and uncertain" (1991, p. 176). The justification of facts is not only empirical but also involves the rational or logical. These two approaches to the justification of factual or propositional knowledge gave rise to a debate between rationalists and empiricists (Pojman, 1998).3 In this section I examine the debate between them, especially its relationship to the justification of biomedical knowledge. Although the empirical is the means by which biomedical scientists justify medical knowledge, objective knowledge is often considered to be rational or logical. However, before engaging the material of this section, I need to discuss first the epistemic conditions for knowing.
According to the traditional definition of propositional knowledge, someone (S) knows a proposition (P) if and only if: (1) S believes P, (2) P is true, and (3) S is justified in believing P (Pojman, 1998). There are three conditions, then, which must be met for knowledge to be propositional. The first is the belief condition, which states that if something is known then a person-or more importantly an epistemic community-must believe it to be the case.' It is a necessary condition that if something is the case, then it must be believed that it is the case. For it would be unusual for a person or an epistemic community to claim knowledge of something but not believe it. However, this condition is not sufficient since something can be believed but not known. Thus, belief pertains to propositions upon which one is willing to place one's or a community's faith that it is the case. That faith is often founded on the metaphysical presuppositions that a person or an epistemic community agrees upon for investigating the world (Collingwood, 1998). Finally, belief is generally contrasted with opinion, which is often based not on evidence but simply on intuition.
The second condition for knowing is the truth condition, which states that if a person or an epistemic community genuinely knows something then it must be true. The consequence of this condition is that something cannot be known that is false. Although belief can be false, knowledge cannot be. In other words, knowledge is necessarily true. Of course, the question arises as to what is meant by truth. Philosophers subscribe to various notions of truth. The most common is the correspondence notion of truth. Proponents of this notion claim that truth pertains to propositions that correspond to the facts. In other words, a belief matches one-toone with the way the world or reality is. Another popular notion is the coherence notion of truth. This notion states that a proposition is true if it coheres with other well known and accepted true propositions. Whether a proposition is true depends on whether it fits in with other known truths or facts. Next, the pragmatic notion of truth states that a proposition is true if it is practical or useful to believe it is true. Truth, according to pragmatists, is what works, especially in terms of ultimately satisfying the knower in a practical manner.
Finally, the emotive notion of truth claims that truth depends upon our emotions or attitudes. This notion of truth may give the impression that all truth is relative, in that a proposition may be true for me but not for you. In other words, there is no consensus as to the criteria for determining a true belief in an absolute sense. There are two problems with this notion. First, truth is not really relative but subjective, i.e. I believe this proposition because I want to. This leads to the second problem: for a person or an epistemic community to know something there must be some type of evidence or warrant for making the claim that a proposition is a true belief. This, in turn, leads us to our final condition for knowing.
The last condition for knowing is the justification condition. Having a true belief is inadequate for saying a person or an epistemic community knows something. The most important epistemological question for any person or epistemic community is: how does a person or a community know that what is known is really the case? This question is about the justification or the proof of what a person or an epistemic community knows. It is the central epistemological question, particularly in the philosophy of science, since Hans Reichenbach (1938) separated, in the first part of the twentieth century, the context of discovery from the context of justification. The answer to this question among philosophers of science has evolved considerably since Reichenbach separated the two contexts.'
The epistemological question is also, of course, an important one for the biomedical sciences: "How can therapeutic claims be justified? What means should be used to instill firm belief in a therapeutic claim?" (Christensen and Hansen, 2004, p. 73). Of course, there is an equal concern over the justification of diagnostic knowledge: how does a clinician know that a patient is suffering from this and not another disease? Traditionally, there are two approaches to justification of propositional knowledge: rationalism and empiricism, to which we now turn especially in terms of medical knowledge's justification.
6.1.1 Rationalism and Empiricism
Rationalists, such as Plato and Descartes, argue that knowledge, especially analytic or a priori knowledge, is known innately or intuitively through the mind's action, although it may never be directly experienced (Pojman, 1998). In fact, rationalists avoid sensate knowledge since it is easily corrupted and the senses are easily deceived. According to Plato, for example, knowledge is achieved by overcoming the world of becoming and the particular in order to grasp the ideal and universal. This innate or intuitive knowledge can then be used as first principles by which additional knowledge can be deduced. Such knowledge is absolute or certain and universal in a sense of being true for all places and for all times, as well as for all people. The propositions that make up this fount of knowledge are often self-evident. As Descartes argued, such knowledge must be clear and distinct and the human mind must have the capacity to know it. Thus, the justification of knowledge for rationalists is strictly a rational or logical affair.
As noted above there are several types of rational knowledge, including intuitive and innate knowledge. Intuitive knowledge depends on rational insight into the phenomenal nature of reality. In other words, it depends upon human capacity to grasp knowledge, especially mathematical knowledge, simply by rational means. Innate knowledge, on the other hand, is constitutive of human nature. Such knowledge is there at birth and is elicited by experiences. This is not to say, however, that the content of the experience is responsible for the knowledge. Both intuited and innate knowledge depend on an epistemic foundationalism in which truth can be directly intuited or is innate to the human condition and thereby forms the basis of additional certain knowledge through deduction.
Empiricists, on the other hand, such as Hume and Locke, argue that knowledgesynthetic or a posteriori knowledge, that is-is obtained only through sense experience. For example, Locke viewed the mind as a tabula rasa or an empty slate upon which knowledge is written through experience. Thus, sense experience provides the justification for knowledge. In the mid twentieth century, Cornelius Benjamin (1897-1968) provided an insightful working definition for empiricism: "Empiricism is that theory of knowledge which holds that descriptive symbols (I) are meaningful, (2) are defined ostensively in terms of hard data, (3) and refer to hard data" (1942, p. 498). "Hard data," a phrase borrowed from Russell, refers to "the clarity with which a datum is given" (Benjamin, 1942, p. 497). For example, a "red spot" immediately sensed or clearly evident is a "hard" datum while myself or a universal represents a "soft" datum. Using the above definition, Benjamin divided empiricism into three types: positivism, factionalism or constructivism, and realism.
According to Benjamin, positivism-especially a "pure" variety-adds two further propositions to the above definition: "(I) All other symbols (e.g., suppositional symbols) are meaningless; (II) Soft data cannot be known to exist" (1942, p. 498). Although the logical positivists and empiricists are certainly positivists to some extent, their position is not a "pure" positivism vis-d-vis the two additional propositions. Fictionalism or constructivism, which Benjamin attributes to Mach and Pearson, adds the following propositions: "(I) Suppositional symbols (1) are meaningful, (2) are defined by operations of construction on hard data, and (3) refer to nothing; (II) Soft data cannot be known to exist" (1942, p. 499). Finally, realism, which Benjamin designates "realistic empiricism" and attributes it to Russell, Whitehead, and Meyerson, adds the following propositions: "(I) Suppositional symbols (1) are meaningful, (2) are defined by operations of inference upon hard data, and (3) refer to soft data; (II) Soft data can be known to exist" (1942, p. 499).
Empiricism is also supported by the "new experimentalism" that arose over the last several decades, from work by historians and philosophers of science. Traditionally scientists depend on experiments and the evidence obtained from them to justify scientific theories. In a review of the early work in the new experimentalism, which focuses primarily on Allan Franklin's The Neglect of Experiment as well as Peter Galison's How Experiments End, Robert Ackermann (1989) drew attention to the "experimental sequences" that these authors rely upon to examine experimental practices in physics. In a critique of the new experimentalism, Deborah Mayo claimed that its proponents ignore or devalue the role of statistical methods in experimentation and she proposed to combine "a standard error statistical tool (significant tests) together with an experimental narrative [provided by the new experimentalists] ... to articulate the procedure for distinguishing artifacts in an important class of cases [e.g. Galison's notion of how experiments end]" (1994, pp. 277-278). Mayo (1996) then developed an "error statistical" philosophy in which hypotheses are linked to evidence through a piecemeal testing that is ultimately ampliative.
Recently, Marcum (2007) has proposed a notion of experimental series that shares certain features with Mayo's philosophy of experimentation: the piecemeal approach to testing, reduction of error by increasing severity of testing, and an ampliative nature of inductive inference. However, the notion of experimental series is not focused on statistical methods of experimental practice per se but on the connection of experimental evidence and not just its conglomeration for justifying a theory.
The debate between rationalists and empiricists also has a long tradition within medicine. For example, the issue of modern theory (rationalism) and conventional practice (empiricism) for medicine was vigorously debated among physicians in the seventeenth century (King, 1978). Giorgio Baglivi, in the face of contemporary rational medicine called for an empirical medicine:
The two chief Pillars of Physick are Reason and Observation: But Observation is the Thread to which Reason must point. Every Disease has, not a fictitious, but a certain and particular Nature, as well as certain and peculiar Principles, Increase, State and Declination. Now, as all these are brought about independently of the Mind, so in tracing their Nature we have no occasion for a subtile and disguis'd way of Disputing, but only for a repeated and diligent Observation of what happens to the several sick Persons, and such an acuteness of Mind as is conformable and obedient to Nature's Measures (1723, p. 9).
It took several centuries after Baglivi until empirical medicine became the standard for medical practice. However, rationalism is also an important epistemological component of the biomedical model; but, "rational therapy can only claim to be true if the theory encompasses all the relevant elements of the disease in question" (Christensen and Hansen, 2004, p. 74). Given this restriction, much of modern medicine's epistemology is driven by empiricism and its attendant technology.
The epistemology of the biomedical model, then, is one of empiricism, not only in terms of methodology but also with respect to the technology that supports an experimental method-for medical knowledge and practice within the biomedical model rely on the technological developments in the natural sciences, especially the physical sciences.' The acquisition and implementation of medical knowledge reflects the techniques and procedures of these sciences. Moreover, the randomized, doubleblind, placebo-controlled clinical trial is considered the "gold standard" for determining the efficacy of a pharmaceutical drug or of a surgical procedure.' "The development and increasing acceptance of randomly allocated controlled clinical trials represents," according to Tobias and colleagues, "...the greatest advance this century in medical technology... we all stand to gain from improvements in treatment validation that cannot reliably be obtained by any other methodology" (Tobias et al., 2000, p. 1371). Such clinical trails and other testing procedures became the foundation for evidence-based medicine (Sackett et al., 1996).8 These scientific practices define acceptable medical knowledge and practice within the biomedical model.
The debate between empiricists and rationalists, however, did not led to a resolution of the problems associated with justifying biomedical knowledge. As Baglivi pointed out centuries ago, both approaches to knowing are critical for the practicing physician:
Those who oppose Reason to Experience, whether Empiricks or Rational Physicians, seem to be all Mad: For how can we make Reason to act all the Parts of Science, that, as all wise Men ought to acknowledge, is acquir'd by Tryal and Use continu'd thro' a long progress of Time? And, on the other hand, why should Experience be only regarded, and Reason turn'd out of doors?...I understand that Queen Reason, that is plac'd above all the rest, by which a Physician looks into the Principles and Causes of Diseases, foretells their progress and event, and gathers Futurities from what's present (1723, pp. 7-8).
The issue arises whether there is a possible synthesis between the two epistemic positions.
Jan van Gijn has proposed an "empirical cycle" for the generation and justification of medical knowledge. "Pathophysiological reasoning leads to hypotheses," according to van Gijn, "while the content of the rational process is to a large extent driven by the results of the laboratory experiments. The hypotheses should lead to clinical trials and the results of these trials, added to newly gained insights in pathophysiology," he concludes, "give rise to new hypotheses for clinical reasoning" (2005, p. 75). In other words, the generation of medical knowledge is the continuous process in which empirical results give rise to theoretical insights that are then subjected to further experimental testing, and so on.
The synthesis may also be articulated in terms of inductive and deductive logic. For example, biomedical investigators propose different theories to account for various medical phenomena. These theories are always undergoing tests as investigators conduct experiments. In other words, a prediction is deduced from a given theory and if the prediction is verified, then the theory continues to be used to guide investigations. However, if anomalies, i.e. observations not predicted by the theory, are observed, then a new or modified theory may be formulated based on the anomalous observations. This new or modified theory is then tested experimentally, and if successful may replace the older theory.
The rationalist-empiricist synthesis may also be articulated in terms of sensory or experiential and theoretical activities connected through cognitional processes. As the empiricists claim, sensory data and observations are the key, if not the beginning, of knowing. But as the rationalists claim, such evidence does not constitute knowing but only evidence. A cognitional process must intervene, in which the relationship among the various data and observations yield an insight into the meaning of the evidence. Based on this insight a theory is then formulated, in order to explain the phenomenon that yields the evidence. Of course, all evidence is theory-laden but to varying degrees-from anomalous evidence to evidence to test a prediction.
Although the biomedical model provides important methodological tools for obtaining medical knowledge and for practicing medicine, there is still much work required empirically and rationally, as well as philosophically, to resolve the epistemological issues facing it. "A lot remains to be done," according to Liberati and Vineis, "in order to create a better understanding of the nature of proof, evidence, and uncertainty; a more balanced research agenda; more coherent mechanisms to improve quality of care; and more substantial cultural efforts to empower patients and consumers" (2004, p. 121). From a rational perspective, a lot of the development depends on what Edmond Murphy calls the "logic of medicine," the topic of the next section.
6.1.2 Logical Reasoning
Although empirical, especially experimental, procedures are the predominant means for justifying medical knowledge in terms of the biomedical model, rationalism in medical epistemology is not completely without importance or impact. Epistemic claims in the biomedical model depend or should depend, especially for their validity and soundness, on the logical relationship of propositional statements obtained from laboratory experiments and clinical trials. For example, diagnosis and treatment of a patient's disease state depend upon step-by-step, coherent (inductive and deductive) reasoning, from assessing the patient's symptoms to determining the appropriate therapeutic modality. Moreover, logical reasoning in medicine helps to fill in the gaps left by empiricism (van Gijn, 2005). For example, in The Logic of Medicine Murphy provides a procedure "for manipulating ideas in Medicine, systematic in the sense that they can be stated formally and subjected to cogent criticism" (1997, p. 9). The logic of medicine, then, is concerned with the analysis of medical data and observations and not just with the relationship of propositional statements.
Logical reasoning is particularly important for interpreting empirical facts. "Reasoning," according to van Gijn, "is required even in the interpretation of clinical trials. Facts cannot always speak for themselves" (2005, p. 74). Indeed, facts are not equivalent to empirical data or observations; rather, they are interpreted experimental data and observations (Lonergan, 1992). In other words, the researcher must have an insight into the relationship among the empirical data as to their intelligibility. That intelligibility is not an empirical object that can simply be grasped by empirical means. For Murphy (1997), "rules of evidence" are critical in the interpretative process for generating factual, objective knowledge. These rules form not only a logical or rational canon for manipulating the relationship of propositional statements and facts but also a hermeneutical canon required for assigning meaning and significance to medical data and observations.
Rationalism, in terms of the logic of medicine, is also important in terms of planning new experiments in order to test theories or hypotheses (van Gijn, 2005). A new trial is expensive and must first make sense in terms of previous biomedical theories and facts. The type of logic associated with the generation of new experiments is deductive.' A new hypothesis or theory is used to predict an observation, which is subsequently tested experimentally. This approach is called the hypothetico-deductive method. If the theory or hypothesis passes the test, i.e. the predicted observation occurs, then the theory or hypothesis is said to be verified (logical positivists), confirmed (logical empiricists), or corroborated (Popperians). However, if the theory fails the test, i.e. the predicted observation does not occur, then the theory is falsified or more often modified.
Unfortunately, the process of verification or falsification is not so straightforward since neither can be absolute; for the theory being investigated cannot be tested directly, because the assumptions behind it form an interconnecting "web of beliefs" (Quine and Ullian, 1978). Moreover, falsification is not so straightforward since scientists may formulate ad hoc hypotheses to rescue an embattled theory (Lakatos, 1970).
22.214.171.124 Frequentist Statistics
In the biomedical sciences the fit between a hypothesis and an experimental or a clinical observation is often not quite as straightforward as in the natural sciences, even with the above problems, due to error on the part of the investigator or variability of the natural phenomena. In the biomedical sciences, the significance of the fit is generally determined through statistical testing and analysis. Murphy defines statistics as the "[s]tudy of inferences from finite samples about random processes and their specifications" (1997, p. 468).
Statistical testing can be either descriptive or inferential (O'Brien et al., 1989). In descriptive statistics the researcher describes a population's characteristics, while in inferential statistics the researcher designs a study in which observations are made from a sample of the population under study. Traditional or frequentist statistical tests, such as the Student's t-test or the x2-test, allow the researcher to determine whether the inferred conclusion is warranted. Statistical reasoning, then, represents a potent means by which to justify conclusions concerning medical knowledge.
The frequentist approach to statistical analysis involves the comparison of two groups, especially in terms of a pharmaceutical drug or a surgical procedure, with one group representing the experimental group and the other the control group.L° The question is whether the difference between them is real or significant or simply due to chance, in terms of experimental manipulation. To determine whether the difference is significant or not, medical researchers conduct statistical tests to obtain a probability value (P value), which gives them confidence about the difference.
The first step in this process is to form a null hypothesis, along with an alternative hypothesis. A null hypothesis states that there is no significant difference, while the alternative hypothesis states that there is. For example, if medical scientists are testing the efficacy of a drug the null hypothesis claims that there will be no difference between treated and untreated groups vis-n-vis the drug, while the alternative hypothesis claims that the treated group will fair better because of the drug, e.g. cancer remission, than the control group. Once the data is collected, the researchers run a statistical test to determine whether the results are statistically significant, i.e. whether the null hypothesis is rejected." If the null hypothesis is rejected then the alternative hypothesis, i.e. the difference between the two groups is real or significant and the drug is efficacious, is accepted by default.
In frequentist statistical analysis, the medical scientist or clinician is concerned with removing error that can influence the interpretation of results, from null hypothesis testing. There are two types of error. In Type I error, the null hypothesis should be accepted, i.e. there is no difference between the two groups, but the statistical test misleads the research into rejecting it. This type of error represents a false positive in that the difference between the treated and untreated groups is not really statistically significant. In Type II error, the null hypothesis is in fact false but the statistical test misleads the research into thinking it is true. This is a false negative in that the difference between the treated and untreated groups is really statistically significant. Type I error is more egregious than Type II error in that the former type of error can result in harm to a patient, e.g. treating with a drug that is not efficacious, while in the later type of error the researcher has missed the effect.
There are several problems with the frequentist approach to statistical analysis of clinical results. First, frequentists do not provide direct proof that the alternative hypothesis is true. "Unfortunately," according to Lewis and Wears, "there may be many alternate hypotheses different from the original one that might have been accepted based on this evidence had they been proposed" (1993, p. 1330). Thus, a P value pertains not to the truth of an alternative hypothesis but only to the null hypothesis. In other words, because there are many alternative hypotheses one cannot be certain that the stated or tested hypothesis is true since it is considered true only by default.
Another problem is that frequentist statistical analysis is concerned with a population and not with an individual, whereas a physician is often concerned with an individual patient. This statistical approach "denies meaning to the assignment of probabilities to single events or hypotheses. Probability assignments are to classes, not to individuals. Thus," conclude Daniel Albert and colleagues, "questions such as `What is the probability that this patient will die tonight?' and `How likely is that diagnosis?' do not make any scientific sense in this view. We can only legitimately ask, `What proportion of the class of patients like this one will die tonight?"' (Albert et al., 1988, p. 64). Although frequentist statistics are very helpful for interpreting research results from large clinical trials, they are for the individual patient "profoundly unsatisfying" (Montgomery, 2006).
126.96.36.199 Bayesian Statistics
Besides frequentist statistical analysis, many biomedical scientists utilize Bayesian statistical analysis to determine the significance and meaning of experimental and clinical results (Broemeling, 2007; Kadane, 2005; Tan, 2001). This analysis is based upon a theorem named after its originator, Thomas Bayes, an eighteenth century nonconformist cleric (Dale, 2003). Lewis and Wears identify two important differences between frequentist and Bayesian statistical analyses: "the nature of the probabilities that we are trying to estimate from the data and the way in which we use the data to modify our estimates of those probabilities" (1993, p. 1329).
Bayesians take probabilities to be an estimate of an event's certainty rather than its frequency. Instead of relying on large numbers or sampling of an event to obtain a rate at which it occurs, the Bayesian estimates the event's occurrence on prior experience. For example, whether a patient responds to a particular treatment is initially based on a researcher or clinician's subjective estimate and experience. Bayesian use of data is conditional rather than unconditional, in terms of determining the truth of the hypothesis. "Bayesians deal with the probabilities of hypothesis, given a set of data," according to Lewis and Wears, "whereas frequentists deal with the probabilities of data sets, given a hypothesis" (1993, p. 1329).
Bayesian analysis is concerned with the relationship of present data with past data, i.e. "how new evidence can be systematically combined with old to maintain coherently the current state of the evidence" (Murphy, 1997, p. 204). In other words, besides present evidence prior evidence is taken into account to determine the probability of a future event. The first step in Bayesian analysis is to assign a probability distribution to an event's occurrence based on prior data. Next, data are collected on the event's occurrence, and these are used to revise the prior probability distribution. For the Bayes theorem allows the combination of a prior probability distribution with present data to generate a posterior probability distribution, which is then used to estimate the probability of a future event's occurrence and to determine its meaning or significance.'
An example in terms of diagnosis may help to clarify the principles of Bayesian analysis (Sahai, 1992). An attending physician wishes to diagnose patients entering an emergency ward in terms of the probability of acute appendicitis (AA), acute pancreatic (AP), or non-specified abdominal pain (NSAP). The prevalence rates for these conditions are as follows: 30% for AA, 5% for AP, and 65% for NSAP. A rebound tenderness test can be used to demarcate between the three conditions. The test consists of pressing down slowly on the patient's abdomen and releasing quickly, which may be accompanied by sharp pain at the site of peritoneal irritation. Previous studies reveal that 80% of AA patients, 15% of AP patients, and 20% of NSAP patients, exhibit rebound tenderness. The posterior probabilities are easily calculated for each condition: 0.64 for AA patients, 0.02 for AP patients, and 0.34 for NSAP patients.13 Consequently, AA is the most probable or likely diagnosis for a patient exhibiting rebound tenderness.
Bayesian analysis provides several advantages over frequentist statistical analysis. First, it is more consistent with the practical reasoning conducted by clinicians: "the Bayes method provides confidence intervals on parameters and P-values on hypotheses which are more in line with commonsense interpretations" (Congdon, 2001, p. 1). Moreover, its prediction of a future event is more precise since it incorporates past information into the determination of that event. Frequentist statistics rarely include such information. Bayesian analysis affords a more dynamic and adjustable statistics. Another advantage is that it provides important information for the practicing clinician concerning the efficacy of a treatment vis-n-vis another competing treatment.
In addition, Bayesian analysis permits an investigator in a research trial to examine the data without subjecting the trial to an increased error rate, as in a frequentist trial. "This is a strong argument for its use in clinical trials," according to Lewis and Wears, "because it may be possible to terminate the trials earlier, thus exposing fewer patients to ineffective or harmful therapy" (1993, p. 1335). Finally, another advantage is that Bayesian analysis incorporates the plausibility of a particular event, e.g. of a therapeutic procedure. The likelihood that a drug or surgical protocol is successful must cohere with other successes or failures of other similar therapeutic procedures, current biological and medical knowledge, and the experience of the individual clinician.
6.2 Subjective Thinking
Although biomedical knowledge, especially in terms of laboratory data and clinical observations, is an important and even a necessary component in medical practice, it is not sufficient, according to humanistic or humane practitioners. What is needed is personal knowledge of the patient. According to Cassell, for example, "when we are sick we do not need impersonal knowledge; we require personalized knowledge" (1991, p. 133). For Cassell and other humane practitioners, the exclusive pursuit of impersonal knowledge hinders the physician from obtaining the personal knowledge that is critical for treating this patient.
Personalized or subjective knowledge is often the information that is ignored or bracketed in scientific medicine; however, it is critical for the patient's healing. The humanistic models of medicine permit "physicians to elicit information from deep within the patient and combine it with objective findings" (Davis-Floyd and St. John, 1998, p. 97). Such information goes beyond the laboratory data to include what Robert Smith calls "human data" Such data involve "information that the patient communicates in words or through nonverbal but uniquely human modes of expression" (Smith, 1996, p. 98).
The problem with the biomedical model, for humane practitioners, is that the physician no longer interacts with an individual patient or that patient's unique circumstances but with the abstract generalities of a patient's disease obtained from statistical analysis of other patients with a similar disease. To reverse this trend these practitioners seek information that is not limited to just a patient's disease state but that also includes information about the person who is suffering from a specific illness. In the biomedical model both laboratory and clinical techniques generate the necessary data needed to identify the disease and to treat it, whereas in humanistic or humane medicine information about the patient as person is also required to treat successfully the illness and to alleviate the suffering associated with it.14
According to Cassell, "three kinds of information about sick persons-brute facts, moral, and aesthetic-are necessary to the work of the clinician" (1991, p. 178). While brute facts about a patient's disease state are required for practicing medicine, they alone are inadequate for the patient's healing. Both the patient's moral values and aesthetic sensibilities are required to understand and treat a patient's illness and to relieve the suffering associated with it. Only when a physician is informed about these values and sensibilities, can he or she genuinely care for a patient and assist that patient on the road to healing. "Information about the patient that is being acquired, evaluated, and utilized and which enters into the value and aesthetic assessments may also include," for Cassell, "feelings, body sensations, and even the spiritual (transcendent)" (2004, p. 226). It is this information that cannot be bracketed from the objective clinical data and observations, which is needed to heal this sick person. Such information is obtained through subjective thinking.
Subjective or personal reasoning or knowing is shunned in science because it is thought to distort universal and objective knowledge, which is considered the only true knowledge. However, that knowledge is personal, according to Michael Polanyi (1962), because its acquisition and justification depend on our unique perspectives, which include, e.g. our intuitions, values, and aesthetics. Polanyi's notion of personal knowledge depends on what he called the "tacit component" of intelligence, a prelogical phase of knowing that is not necessarily articulatible. It is this component that not only allows for the acquisition of knowledge but also the means to determine its meaning, especially for the human knower.
Objective knowledge is only part of the story for understanding the world, the other part is what that information means for a particular knower. Polanyi rejected the fact/value dichotomy and provided the necessary scaffolding for the current development of emotional intelligence (Tauber, 2008). His personal knowledge prepared the way for other epistemological projects, especially humanistic or humane medicine. Two of these projects-conducted by Foss and Tauber-are briefly discussed below, after which several components of personal or subjective knowledge-including intuitions, values, and virtues-are examined in further detail, followed by a discussion of narrative reasoning.
Laurence Foss (2002) proposes an "infomedical" model in which information, especially in terms of the psychoneurological, can be incorporated into medical practice. His main thesis is that the self as body and mind must be reintegrated as a unit into medical practice. He uses the infomedical model to argue for a "holistic science of self-referentiality" (Foss, 2002, p. 70). Instead of viewing matter as res extensa and causation as strictly upward, he argues for viewing matter as res autopoietica and causation as mutually upward-and-downward. According to the infomedical model, the mind and body are connected by information. To adapt this model to the clinic requires a mind-body dictionary based on "intersemiotic transduction," in which, for example, information is sent from the mind (sender) to immune cells (receiver) via neurohumors (channel). Thus, the mind-whether conscious or unconscious-can influence a patient's health.
Foss also put forward a mechanism for information transfer among parts of the organism, as well as between the organism and its environment. By infusing matter with conscious properties he reformulates the second law of thermodynamics as the second law of psychothermodynamics, in which "the universal dynamic is vitalistic and autopoietic" (Foss, 2002, p. 233). Finally, Foss converts objectivity to a subjectified objectivity: "the object is a subject, the patient is an agent, each possessing some limited degree of autonomy" (2002, p. 242).
With these changes in place Foss attempts to revolutionize or humanize medicine through a relational model of biology, in which additional information about a body part is determined from that part's context within the organism and its environmental context. It is this information that allows an organism to reform itself in response to external challenges, such as disease. Foss' infomedical strategy is that "the organism as a whole exhibits mindful self-regulating behavior" (2002, p. 269). Thus, for humane practitioners subjective knowledge-of how the patient interprets the experience of illness and provides meaning for it-affects how the patient responds to the illness and its treatment.
Tauber proposes a model for medical knowing that joins together the objective and subjective ways of knowing, through the knowing subject. His proposal is based on a study of Henry David Thoreau's attempt to correct the objectification of knowledge in the nineteenth century, due to the rise of positivism. "Radical objectivity fails because," according to Tauber, "the view from nowhere leaves Man out of the picture, and with no perspective there is no significance, no meaning, no order, and ultimately no self' (2001, p. 21). In the quest for objectivity, the self or subjectivity in terms of the knowing subject is abandoned along with important moral characteristics and values that guide knowing. This has a major impact on modern medicine, which exchanged its empathic character for a dispassionate one. For Tauber, "the glue holding together the various epistemological strands of contemporary medicine is of a personal moral character" (2005, p. 10). In other words, what he is seeking is a rejoining of fact and value that the positivists tore asunder.
Tauber's calls his proposal "moral epistemology-moral, because clinical evaluation and care are value-laden, and epistemological, because medicine expresses and employs the form of knowledge" (2005, p. 9). Facts are always given within a context of values and are thus products of interpretation; for values influence and guide knowing. By salvaging facts from positivist objectification, Tauber opens up a space in which to incorporate a patient's values with those of the medical profession. The moral imperative of medicine, then, is to identify a patient's subjective values in order to situate a patient's objective clinical facts. Physicians must become more self-reflective morally and must integrate this moral reflectivity into the technical demands of medicine. Tauber makes several recommendations to that end, such as including an ethical section in the medical record. Such subjective ethical and moral knowledge would complement the objective and scientific knowledge to yield a more comprehensive picture of the patient.
Although humanistic or humane models share many epistemological features with the biomedical model, e.g. the assumption that logic is important for practicing medicine, they also rely to some extent on the humanistic practitioner's intuitions. "Intuition in medicine," according to Irvine Page, "is crucial" (1978, p. 218). It is a critical skill of a "good" physician." Intuitions are not necessarily impediments to sound medical judgment and practice; but when judiciously utilized and constrained by the epistemic and empirical boundaries of the biomedical model, they enable a physician to evaluate information about a patient's illness that may surpass quantified data, e.g. laboratory test results. This information obtained from a practitioner's use of intuitional, unconscious resources is not just objective or quantifiable but also human-for behind such information is the face of the "Other" (Tauber, 1999). Such information is important for practicing the art of humane medicine.
What is intuition? William Davidson has provided two definitions. The first is an etymological definition: "the apprehension or discerning of a thing actually presented to the eye" (Davidson, 1882, p. 304). It is based on the literal meaning of the word from its Latin root. The second is philosophical in nature: "an immediate perception of the external object seen" (Davidson, 1882, p. 305). Intuition pertains not only to objects in the world but also to ethical qualities and cognitive principles. Moreover, besides the criterion of immediacy, which can be either independent or temporal in nature, two other criteria of intuition include the universal and the irresistible. Intuition is universal in terms of "not admitting of exception," while irresistible refers to the power of attraction (Davidson, 1882, p. 308).
In contemporary philosophy, intuition is a way of knowing in which a person qua mind immediately apprehends an object, a phenomenon, a decision, or a solution to a problem, without any intervening conscious, cognitive processes." Trisha Greenhalgh lists several features of intuition, including "rapid, unconscious process, contextsensitive, comes with practice, involves selective attention to small details, cannot be reduced to cause-and-effect logic..., [and] addresses, integrates, and makes sense of, multiple complex pieces of data" (2002, p. 396). Intuition is a tacit process that matures, as the practitioner gains more experience. It is also a very creative process that defies simple reduction to an algorithm or set of operational rules, such as inference rules in deductive logic. Finally, intuition is a mental habit of hunches.
Historically, intuition is often contrasted with reason as a competing method of knowing. Reason or reflection is a mediated activity (Davidson, 1882). A conclusion to a syllogism, for instance, is immediately obtained not upon inspection of the major premise but through mediation of the middle term. Intuition, on the other hand, is not mediated by any such process. Moreover, reason is considered a superior way of knowing compared to intuition. According to Davidson, "intuition, standing alone, gives us only `an obscure and indistinct consciousness'; for a consciousness `clear and distinct,' Reflection is required" (1882, p. 309)."
Miranda Fricker (1995), however, argues that such a contrast is based upon a rather "thin" notion of reason, in which reason is based on a set of criteria or rules. She contends for a "rich" notion that includes intuition in thought processes. Based on Kuhn's use of intuition to account for paradigm changes, Fricker defines intuition "as a non-inferential, typically subconscious mode of hypothesis formation. It constitutes," she continues, "a sub-personal level of cognitive operation that is crucial to rational enquiry, since it is primarily the intuitive mode which enables us to solve new problems in light of the old" (1995, p. 184). In other words, intuition is a skill of the reasoning process needed to formulate possible solutions (hypotheses) to problems. Intuition is reasonable, then, since these solutions or hypotheses are generated not randomly but selectively. Moreover, it is often involved in determining the acceptability of cognitive conclusions. Consequently, Fricker's "rich" notion of reason includes a reciprocal relationship between intuition and "thin" reason, especially with respect to the generation of hypotheses and their acceptance.
According to Greenhalgh, most physicians acknowledge the importance of intuition in clinical reasoning and practice. She illustrates its use from her own practice. After examining an elderly male patient, whose chief symptom was abdominal pain, and finding no unusual clinical signs, Greenhalgh "went home that night and told [her] husband that [she] had seen a man who was going to die" (2002, p. 395). Indeed, the man died four days later from a strangulated volvulus. She interprets her hunch in terms of intuition. "When I predicted his impending death," Greenhalgh concludes, "I was not consciously aware of the intermediate steps that led me to my hypothesis, but when I learnt," she adds, "the outcome and sought a debriefing with his regular GP, the pieces of the jigsaw were revealed to both of us" (2002, p. 399). Other physicians also point to the importance of intuition. For example, Tauber notes that "the science of medicine is so often guided by intuition" (1999, p. 7).
During the first half of the twentieth century, logical positivists claimed natural science to be a value-free enterprise, in order to guarantee the objectivity of scientific knowledge. However, during the latter half of the twentieth century, especially after the historiographic revolution in the philosophy of science, values emerged as important factors in the scientific enterprise. For example, Kuhn (1977) claimed that the justification of scientific knowledge requires the transformation of the objective criteria of accuracy, consistency, fruitfulness, scope, and simplicity, into similarly denoted subjective values, values that influence the justification of scientific knowledge but do not determine it. Justification, then, "requires a decision process which permits rational men to disagree, and such disagreement would be barred by the shared algorithm [objective criteria] which philosophers have greatly sought" (Kuhn, 1977, p. 332).
After Kuhn, science is now viewed as a value-ladened enterprise and its knowledge as value-dependent. For example, Robert Proctor (1991) argues that scientific knowledge is not neutral but rather driven by political and societal values. Again, Tauber (2007) claims that the fact/value separation in science is specious and that science is imbued with values that serve an epistemological function. Humanistic practitioners also acknowledge the importance of values for medical knowledge and practice. Indeed, Cassell claims that "a value-free medicine is a contradiction in terms" (1991, p. 185).
But, what is a value and how is it used epistemologically? The notion of value is not easily defined and there are several approaches to its definition within the philosophical literature. Values are also used in a variety of fashions. Tauber (2005), for instance, distinguishes three uses based on Najder's analysis: quantitatively, in terms of the value of something, attributively, in terms of something being conferred value, and axiologically, in terms of a principle which one uses to assign value. Although the axiological use of values is examined in Chapter 11, in the remainder of this section their use is explored not only in the justification of scientific and medical knowledge but also in its acquisition. Finally, William Stempsey warns about the difference between value and personal preference, especially for medicine: "Personal preferences do play an important role in our ideas about the value of health and disease, but I will argue that there are other objective values that ought to be recognized as values by any person, whether or not that person has a preference for them" (2000, p. 42).
Ernan McMullin (1982) divides the use of values in science into two categories: epistemic and non-epistemic.18 Epistemic values are those that are used to advance the veracity of scientific claims. They are important for assessing a "fit" between scientific theories and the natural world and include, e.g. external consistency, fertility, internal coherence, predictive accuracy, simplicity, and unifying power. Non-epistemic values are those values that can be used, when epistemic values fail to distinguish between empirically equivalent theories. They do not enhance a theory's "epistemic status" but reflect specific cultural, social, political, and religious beliefs. Although these values are influential in the short run within a community of practitioners, they are eventually replaced by epistemic considerations. In a study on the development of evolutionary science during the nineteenth and twentieth centuries, for example, Michael Ruse (1999) demonstrates a shift from non-epistemic to epistemic values in its practice.19
Besides the categorization of values as epistemic or non-epistemic, they can also be divided into factual or ethical in terms of the pursuit of scientific and other kinds of knowledge (McMullin, 1982). A factual value is not limited simply to the absolute correspondence of the world to scientific theories but also to the corrigibility of scientific theories in light of additional evidence. Ethical values are important to a professional community and its proper moral function. Certainly scientists and theologians, for example, share a genuine desire to know the facts and to conform to ethical values that ensure them. For instance, honesty is the disposition to tell not only the truth but also to avoid telling a lie. Moreover, honesty involves uprightness and reliability of character." These values are essential for the acquisition of knowledge in most disciplines. Although scientists are often portrayed as being more objective than those in other disciplines, postmodern studies have deflated that caricature.
Values in medicine serve epistemic and non-epistemic, as well as factual and ethical, functions in the acquisition and justification of medical knowledge, especially for humanistic or humane practitioners. For example, Cassell advocates the need for knowing a patient's values in order to obtain the patient's "personal knowledge" (1991, p. 172). Values, then, are critical for gaining a comprehensive picture of the patient, which is needed for adequately treating a patient: "applying medical science to particular patients mandates thinking in terms of values as much as in terms of the objective facts of the body" (Cassell, 1991, p. 107). Moreover, values are also critical for determining the nature of health and disease.
Cassell (1991) identifies five sources of values needed for medical knowledge and practice. These include the values society places on health and illness, the goals of medical care in general, physicians' personal and professional values, people's individual values, and the values that under gird the operations of a system as a complex unity or whole. "Values, then, like scientific facts, are essential," according to Cassell, "to the clinician's knowledge of sick persons" (1991, p. 184).
Tauber also argues for the importance of values in medical knowledge and practice: "values structure all facts so that their meaning and significance only take form when they are sorted, organized, prioritized, and acted on as determined by the rules governing the value-based choices optimizing patient care" (2005, p. 240). The traditional distinction then between facts and values is a false dichotomy and Tauber proposes to collapse facts and values in terms of a moral (values) epistemology (facts).
To support a moral epistemology, Tauber (2005) divides values into positivist and nonpositivist categories. Positivist values are objective and neutral and guarantee medical knowing as scientific knowing. Although medical knowledge can profit from incorporating these values, exclusive use, however, as in the biomedical model robs medical practice of its humane dimension. That dimension requires nonpositivist values, which are subjective and reflect the personal goals of the patient and healthcare provider. In other words, positivist values are necessary for the physician's knowledge of the patient but are not sufficient, "for the glue holding together the various epistemological strands of contemporary medicine is of a personal moral character" (2005, p. 19). This moral dimension of medicine based on these nonpositivist values is what makes medicine the humane practice that it should be.
Recently the role of virtues in the acquisition and justification of knowledge has gained prominence in philosophy, in a sub-discipline called "virtue epistemology." "The name `virtue epistemology'," according to Linda Zagzebski and Abrol Fairweather, "has come to designate a class of recent theories that focus epistemic evaluation on properties of persons rather than properties of beliefs or propositions" (2001, p. 3). Virtue epistemology is based on virtue ethics, in which actions of persons are analyzed in terms of the normative characteristics of the person rather than of the acts themselves. In like manner, virtue epistemologists are interested in the normative characteristics of the person than in the knowledge itself. As noted above, traditional objective epistemology focuses on knowledge production and justification in terms of the evidence or methods used to produce it, while virtue epistemology focuses on the intellectual virtues of the epistemic agent.
Intellectual virtues are divided into two types (Greco, 2000). The first pertains to the reliable or sound cognitive faculties or capacities, including the senses, especially vision, memory, intuition, inferential reasoning, and introspection, necessary for obtaining and ensuring knowledge. This kind of virtue epistemology is called "reliable virtue epistemology," since knowledge as justified true belief is based on the reliability of cognitive faculties and processes (Sosa, 1991; Greco, 2002). The second type of intellectual virtues pertains to the virtuous features of the epistemic agent, such as honesty, open-mindedness, humility, fairness, curiosity, tenacity, and integrity. This kind of virtue epistemology is called "responsible virtue epistemology," since knowledge is based on the epistemic agent's responsible and conscientious activities (Zagzebski, 1996; Roberts and Wood, 2007).
Although virtue epistemology is not fully utilized in contemporary medical epistemology, the virtues of the physician, whether reliable or responsible, are important both for the acquisition and substantiation of medical knowledge for clinical practice. Not only must the physician's cognitive faculties and capacities function properly, but his or her disposition must be sufficiently responsible to warrant an accurate diagnosis and appropriate therapeutic modality. For example, a physician must be honest in terms of evaluating the clinical data and observations and not allow biases and prejudices to distort their interpretation.
6.2.4 Narrative Reasoning
"Biomedical reasoning may be sufficient to explain the bounded realm of microscopic events and abstract principles, but other kinds of reasoning are necessary," according to Linda Hunt and Cheryl Mattingly, "when those principles are applied to the unbounded universe of the real world of physical, phenomenological, and social lives" (1998, p. 270). One of the more prevalent alternative forms of reasoning to biomedical reasoning is narrative reasoning. In contrast to the objective facts and to their logical analysis associated with objective, biomedical reasoning, the humanistic or humane models incorporate the patient's narrative of the illness experience into medical practice that utilizes subjective and personal reasoning.
Narrative reasoning, for Barbara Schell, "involves thinking in story form" (2003, p. 136). This type of reasoning allows the humane practitioner to access personal information concerning illness' disruption of the patient's life. Its main function is to make sense of the confusion and anxiety illness introduces into the patient's lifeworld. Whereas logical biomedical reasoning is concerned with the validity and soundness of the arguments and the truth of medical statements, narrative reasoning is concerned with the meaning and significance of the patient's illness story. Practitioners of narrative medicine ask questions about the nature of a patient's illness experience, while biomedical practitioners ask questions about the nature of the disease itself.
Kathryn Montgomery also maintains that medical practice should be grounded in narrative reasoning: "Physicians use both the scientific or hypothetico-deductive and the practical or interpretative and narrative, but it is the latter that makes them clinicians" (2006, p. 45). Narrative reasoning is a case-based rationality and involves the interpretation of a patient's illness experience. It is not reducible to a set of inference rules, but requires a hermeneutical canon for interpreting a patient's story. Rather than banning anecdotal knowledge, narrative reasoning depends upon it for making the best possible clinical judgment and decision.
Narrative rationality, according to Montgomery (2006), is akin to Peirce's notion of abduction. Clinicians begin with a particular patient before them and based on the presenting symptom(s) collect preliminary evidence, which they interpret in terms of the patient's narration of the illness experience. Clinicians then continue to collect further evidence based on the patient's story, until the cause of the patient's illness is determined. The process is a "circular, interpretive process" and the information clinicians gather is not a set of isolated abstract facts but rather facts connected through an intricate narrative, both on the part of the patient and the clinician (Montgomery, 2006, p. 47).
Mattingly (1998) identifies three features of narrative reasoning in medicine. The first involves the motives that animate a patient's story, especially in terms of a patient's actions and the consequences of those actions. "In narrative reasoning," according to Mattingly, "an `inner world' of motive and desires is seen as the significant underlying cause of events" (1998, p. 284). For Montgomery (2006), medical causation is best explicated in terms of narrative reasoning rather than in biomedical statistical analysis. Although medicine strives for simplicity in terms of causation as an ideal, the practice of medicine reveals that causation includes, besides the pathophysiological, the psychological and cultural-for illness is expressed at these various levels.
Striving for the ideal of scientific causation misrepresents the true nature of clinical causation. "Because clinical reasoning is retrospective," argues Montgomery, "it needs to be represented in a way that allows a larger, looser concept of cause than linear cause and effect. What is needed," she insists, "is representation that can accommodate time and chance. Narrative," she concludes, "provides for the circumstantiality or (probably) noncontributory detail and leaves room for contingency, conjunction, and multiplicative causes that unfold over time" (2006, p. 80). Although the statistical approach to clinical causation is necessary for a secure foundation to medical practice vis-a-vis biomedical facts, narrative provides access to subtler dimensions of it.
The next feature of narrative reasoning involves the construction of a patient's social world. Narrative reasoning allows a physician to enter into a patient's social world in order to better understand the impact illness has on a patient's lifeworld. "Narrative provides a wonderful vehicle for making sense of actions, because," explains Mattingly, "it seeks to make actions comprehensible by showing how they are responsible from the agent's perspective" (1998, p. 285). It is that perspective that provides a physician with the critical information for addressing a patient's existential concerns, which are an important component of a patient's illness experience and require addressing in order to heal a patient fully. "To know what patients endure at the hands of illness and therefore to be of clinical help," argues Rita Charon, "requires that doctors enter the worlds of their patients, if only imaginatively, and to see and interpret these worlds from the patient's point of view" (2006, p. 9). This type of knowing distinguishes between a biomedical practitioner and a genuine healer (Davis-Floyd and St. John, 1998).
The final feature of narrative reasoning involves the probable and possible rather than the determinant and necessary, as in logical, biomedical reasoning. "Narrative is needed," according to Mattingly, "to contemplate the world in its complexities and to decipher how one should navigate one's way in it, for narrative is built on surprise, chance, contingency, [and] the anomalous event" (1998, p. 289). Narrative reasoning is able to assist a person in navigating life's exigencies and in making sense of them, because it is grounded in the practical or phronetic (Charon, 2006; Mattingly, 1998; Montgomery, 2006). As practical reasoning, it is concerned with the good; and, for medicine, the good is defined in what is best for the patient. Patient care then "requires practical reasoning, or phronesis, which Aristotle described as the flexible, interpretative capacity that enables moral reasoners (and the physicians and navigators he compares with them) to determine the best action to take when knowledge depends on circumstance" (Montgomery, 2006, p. 5).
Montgomery (2006) examines the process of narrative reasoning in terms of maxims, beginning with various rules-of-thumb. She contrasts these informal rules with formal decision analysis prevalent in current academic medical circles. Although these decision procedures are aids to clinical practice, she warns that they are no substitute for it. Moreover, the informal rules or maxims are generally expressed as contradictory pairs. For example, in history taking a physician must balance the maxim that a patient's articulation of the presenting symptom is key to diagnosis with the maxim that one must be wary of whether a patient's articulation of that symptom is accurate or truthful. Although the reliance on contradictory maxims appears undignified for a profession that celebrates its reliance on science, Montgomery insists that the general nature of medical practice demands it. These rules "were never meant for universal application; they are situational wisdom that have arisen out of (and proven useful in) circumstances very like those identified in a particular case" (Montgomery, 2006, pp. 117-118).
Montgomery (2006) also examines maxims that guide a clinical encounter to those that guide a clinical mindset with respect to clinical thinking and judgment. These maxims are metarules or phronesiological maxims, which function at a broader interpretative level. One of the most important ones, according to Montgomery, is the "When you hear hoofbeats, don't think zebras" maxim. This maxim "reminds clinicians that the presence of signs and symptoms shared by a number of diagnoses is not likely to indicate the rare one on the list" (Montgomery, 2006, p. 122). However, a clinician must also be aware that a patient's symptoms may point to a rare disease. Moreover, there are other maxims that govern clinical thinking and judgment. For example, in terms of the goals of medicine the contradictory pair is to do everything possible and to do no harm. She identifies several lessons from this phronetic approach to clinical reasoning, especially the lesson that one should learn from one's elders but question what they teach you.
The biomedical model is patterned after objective, scientific thinking and reasoning. It is concerned with the logical validity of its arguments and the truth or veracity of its propositional knowledge. In contrast, humanistic or humane models are patterned after subjective ways of thinking and reasoning that include intuitions, values, virtues, and the illness story. Moreover, subjective ways of thinking deal with issues that are often not addressed by objective ways of thinking but nonetheless are important for a patient's wellbeing. As such, these subjective ways of thinking are championed as means to address the alienation and objectification patients feel when treated by biomedical practitioners, and consequently serve to address the quality-of-care crisis.
Finally, Hunt and Mattingly (1998) claim that objective and subjective thinking or reasoning are not contrary to one another, but rather they are complementary. In other words, subjective thinking or reasoning instantiates objective thinking or reasoning. As Lonergan articulates the resolution of the relationship between objectivity and subjectivity from a larger perspective: "Genuine objectivity is the fruit of authentic subjectivity" (1979, p. 292).