The Deficit View As It Is Currently Practiced

The deficit approach to autism as it is currently practiced is problematic in at least three respects. First, those who take a deficit approach to autism tend to focus one-sidedly on deficits, to overlook alternative explanations, and to be too quick to assume that one offers a valid explanation or faithful description of the relevant phenomena merely by pointing to a lack or an absence. For example, consider the research on differences in face perception in individuals with autism. It is a standard belief, at least in the West, that a lot of visual information about people's feelings and thoughts is captured in their eyes. Thus, face perception is taken to play an important role in social cognition. Researchers investigate autists' looking and face scanning behavior while perceiving a face to try and determine whether and how autists' face perception differs from that of comparison groups.

Some studies suggest that autists look at faces for less time and look at different parts of the face than non-autistic individuals (Falck-Ytter and Hofsten 2011). People with autism tend to look at the other person's mouth or chin rather than their eyes. In a well-known study Klin and colleagues (2002) recorded autists' scanning behavior while viewing an emotional scene of the film "Who's Afraid of Virginia Wolf?" and found that autists spent more time looking at the actors' mouths than the actors' eyes.

The dominant explanations for these behavioral differences are deficit-based. 7 For example, it is hypothesized that people with autism do not attend to the most emotionally salient region of faces because they are not interested in social stimuli (Grelotti et al. 2002) 8 and that when they do look at the most emotionally salient region of faces perceptual processing deficits specific to emotions and faces compromise their ability to see the emotional significance of faces (e.g., Klin et al. 2002; Riby and Hancock 2009).

A viable but neglected alternative interpretation of autists' lack of eye contact and atypical face scanning behavior is that some autistic individuals actively avoid eye contact and do not look at faces as often or for as long or at the same face regions as non-autistic individuals because it is overwhelming and disturbing for them to do so (Cole 1999). Some autists may experience the "overwhelming presence of another" (Cole 1999, 90) when they meet another person's gaze. Autists may also have a stronger skin conductance reaction to direct gaze (Tanaka and Sung 2013) and heightened activation of the amygdala, a part of the brain that plays a key role in processing emotions, during gaze fixation (Dalton et al. 2005). These findings suggest that autists may have a heightened emotional response to direct eye contact.

Many autobiographical accounts of autism support this second line of explanation. For instance, Donna Williams, an autistic adult, wrote in her best-selling memoir, Nobody Nowhere (1992), that she was overwhelmed by perceiving faces. Here she explains how she developed ways to avoid looking at them: "I had recognized three ways in which I had avoided looking at people. One was to look straight through what was in front of me. Another was to look away at something else. The third was to stare blankly ahead with one eye and turn the other one inwards…blurring whatever view I had" (Williams 1992, 173, quoted in Cole 1999, 91). The second line of explanation casts doubt on the claim that seeing is dissociated from feeling in autistic individuals.

A second way the deficit approach to autism is problematic, as it is currently practiced, is that there is moral complacency in how deficit attributions are made. Deficit attributions require a standard by which to measure deficiency or inadequacy, and this requirement brings with it the danger of moral complacency in how the value judgments that inform deficit attribution are made. Value judgments are made automatically and unthinkingly. Suppose for the sake of argument that the neurological, psychological, and behavioral lack attributions to autists are true. What legitimates the move from attributions of mere lack to attributions of deficits to autists in these domains? What makes it the case that autists' amygdala ought to be such-and-such size, that autists ought to make eye contact, for example? There is a seamless move from attribution of lack or absence to attribution of a deficit. The difficult and crucial question of what makes it the case that X ought to have or be F oftentimes is not explicitly addressed.

A closely related manifestation of moral complacency in deficit attribution is the tendency to assume that the values that inform deficit explanations as they are currently practiced are the values we should be using to make judgments about observed lacks and absences in people with autism. Accepted values may be so entrenched that those making deficit determinations do not even recognize that they are making a choice to assign negative value to the deviation.

The conflation of autism with suffering is a case in point. Murray (2012) observes that in the dominant scientific and non-scientific narratives of autism, "suffering is integral to the manifestation of the condition and subsequent diagnostic process…it is one of the ways by which autism is properly known" (21). However, many people labeled autistic do not experience autism in terms of severity and suffering. And many parents of children with autism find it misleading to conceptualize their children as "autism sufferers" or as "suffering with" autism. A revaluation of values is needed here. There is no easy equation between autism and suffering or, more generally, between absence and suffering. 9 Instead, the value judgments that underlie inferences from autism and from lack or absence of F to suffering should be discussed as part of an open empirical process that includes the voices of autists whose interests and values are represented in the determinations (Silvers 2003).

Third, there is a tendency to use deficit-like terminology to describe autists in interpretations of differences between autistics and non-autistics when the interpreter knows that the autistic interpretee is autistic. 10 Mottron (2011), an autism researcher and autism rights advocate, provides compelling examples of this tendency when interpreting functional magnetic resonance images (fMRIs) of autistic subjects' brains:

[R]esearchers…systematically report changes in the activation of some brain regions as deficits in the autistic group — rather than evidence simply of their alternative, yet sometimes successful, brain organization. Similarly, variations in cortical volume have been ascribed to a deficit when they appear in autism, regardless of whether the cortex is thicker or thinner than expected. (34)

Mottron (2011) also notes that there tends to be a positive bias in favor of typical individuals when interpreting differences in intelligence test performance in autistic children and typically developing ones. Typically developing individuals are not described as having a deficit when autistic individuals outperform non-autistic comparison groups. Moreover, when superior performance is observed in autistic test subjects it is sometimes explained as a mere byproduct of a deficiency in some other capacity or feature rather than as a strength in its own right. For example, exceptional performance on the Block Design nonverbal subtest and embedded figure tasks are often understood as low-level perceptual strengths that are byproducts of high-level conceptual deficits. 11

Mottron's (2011) critique suggests that whether some observed difference counts as a deficit partly depends on how the person interpreting the observed difference conceptualizes the person that exhibits that difference. In these examples, whether an observed brain difference counts as a deficiency or as some other kind of variation, or whether a test performance counts as a deficit, mere byproduct, or strength in its own right partly depends on whether the person being tested is already understood as autistic.

Like the first and second problems discussed above, the third problem is not a problem with deficit explanations per se but with explanation by deficit as it is currently practiced. It is a tendency one has to be aware of and preempt if one takes a deficit-based approach. To say that concern is warranted here is not to claim that if researchers are looking for deficits, they will find them. We should allow that even researchers who are searching for deficits may end up with findings that contradict their initial hypotheses.

The Very Idea of a Deficit View

It is not difficult to see how the deficit approach to autism, as it is currently practiced, is problematic. There is moral complacency in the choice and application of standards in deficit attribution, unreflective use of "lack" and "deficit" to characterize differences observed in autists' brains and behavior, and a tendency to focus one-sidedly on deficits and overlook alternative explanations. It is harder to make out what it would mean to say that the very idea of a deficit explanation or approach is objectionable, at least for some phenomena. Below I explore two ways to develop this stronger line of objection to deficit views.

First, to conceptualize a person primarily in terms of deficits is de-humanizing. Roughly speaking, de-humanization occurs when individuals or groups are understood, evaluated, and treated as lacking in full humanity. Some forms of de-humanization, like those found in the context of war, slavery, and genocide, for example, are extreme and explicit. Some are more moderate, passive, subtle and implicit (e.g., perceiving people from another culture as less capable of characteristically human emotions) (Bain, Vaes and Leyens 2014). The form of de-humanization of individuals with autism that I examine below is more akin to the second kind, where de-humanized individuals or groups are perceived, evaluated, or treated as lacking some feature, trait, capacity and so forth that is taken to be characteristically human.

Consider the "lack of empathy" portrayal of autism, one of the most common scientific and non-scientific representations of the condition by experts and laypeople alike. It risks de-humanizing autistic individuals. On the "lack of empathy" hypothesis, autists are said to have an empathy deficit whereby they are incapable, to varying degrees, of "putting themselves into someone else's shoes, imagining the world through someone else's eyes and responding appropriately to someone else's feelings" (Baron-Cohen 2003, 137). Autism is increasingly positioned as an empathy deficit while empathy is increasingly positioned as a fundamental human trait both in popular culture and in research on autism (McDonagh 2013). Taken together, these two trends prompt one to conceptualize autists as deficient in a hallmark of being human. 12

De-humanization is used to justify the oppression, exclusion, exploitation, and marginalization of the de-humanized other. These far-reaching, demoralizing effects of extreme and overt de-humanization have been felt across cultures and historical eras. Recent research in social psychology (e.g., Haslam et al. 2005; Haslam 2006; Leyens et al. 2007) suggests that more moderate, subtle, and passive forms of de-humanization have less obvious but nonetheless harmful effects that are not well understood. 13 De-humanization may manifest at the bodily interaction level, for example, when negative beliefs about an out-group compromise mimicry behavior (imitating intonation, posture, and facial expressions) and one's abilities to perceive the behavior of members of the out-group as expressive of their affective states and to intuitively grasp their intentions and actions (Gallagher and Varga 2014).

The above-cited social psychology research investigates the de-humanization of those one perceives as members of cultural, racial, or ethnic out-groups, but arguably the issues extend to other kinds of "others," including those labeled autistic. The possibility that negative beliefs about an out-group can negatively impact one's social cognition capacities (e.g., perception of intentions and motor mimicry) when trying to make sense of an out-group member's behavior raises serious questions about whether and how the "lack of empathy" conception of autism impacts non-autistic individuals' abilities to recognize autists' emotions and intentions, or to perceive them as having a "fully human" inner life when autistics are perceived as out-group members. 14

A second way of reading what it would mean to say that the very idea of a deficit approach is objectionable in the case of autism, and perhaps more widely, is that purely deficit views impede progress in our understanding of the phenomena themselves. "Perhaps more widely", in my above formulation, leaves open the possibility that there could be a purely deficit view of some phenomenon that does not impede progress in our understanding of that phenomenon. "Purely" allows that deficit attributions can be part of a larger explanatory framework that includes attributions of excess and even attributions of different qualitative forms, resulting in an overall explanation that goes beyond conceptualizing the phenomenon simply in terms of what is missing. With these qualifications in mind, below I unpack the claim that purely deficit views' blinkered reliance on privative description and explanation impedes progress in our understanding of the phenomena themselves.

Sass's critique of the deficit view of schizophrenia (2000; 2000-1a; 2000-1b; 2001) furthers reflection on this relatively neglected issue. While Sass does not make the distinction between problems with the deficit view as it is currently practiced and the very idea of a deficit view as objectionable or explicitly discuss the conceptual interrelations between "lack," "deficit," "deviation," and "pathology," his careful analysis of the drawbacks of purely privative description and explanation in the nosology of schizophrenia proves helpful for developing the idea that a purely deficit view impedes progress in our understanding of the phenomena themselves. Below I summarize Sass's critique and distill some general lessons from it concerning the uses of quantitative and qualitative concepts in understanding variation, which I then apply to the case of autism. 15

The purely deficit view of schizophrenia sees schizophrenia "entirely in terms of the loss of higher or more quintessentially human capacities of mind and spirit" (Sass 2000-1a, 56) or in terms of defects or deficits taken to involve "decline of one or more of the higher cognitive faculties widely considered to define the human essence, such as rationality, volition, or the capacity for abstraction or self-consciousness" (Sass 2001, 252). The deficit view finds its expression in the set of features that mainstream American and British psychiatry, at least, increasingly conceptualizes as the defining features of schizophrenia, the so-called negative symptoms, many of which are identifiable by the deprivative "a" that precedes each feature or function that is hypothesized to be lacking: alogia or poverty of speech, affective flattening, avolition, apathy, anedonia, and inattentiveness (Sass 2000).

Sass critiques the deficit view by undermining the positive-negative symptoms distinction on which it relies. A common assumption is that positive symptoms reflect excess and negative symptoms reflect lack. For example, thought blocking is characterized as a negative symptom (a thought is lacking or removed) while thought insertion is characterized as a positive symptom (a new thought is added). The positive-negative distinction may seem commonsensical and straightforward, but "its logic as well as its clinical accuracy are actually quite problematic" (Sass 2000, 154). One problem with the distinction is that there is arbitrariness in classifying something as positive or negative since "[t]he absence of one thing will, after all, inevitably allow for, indeed necessitate, the presence of something else" (Sass 2000, 154). For example, flat affect, a so-called negative symptom, is often accompanied by the presence of something abnormal and anomalous: mask-like faces or incongruous facial expression. Likewise, asociality, the absence of other-directed, socially oriented behavior, is often accompanied by the presence of unusual or socially inappropriate, self-directed behavior (ibid).

Another problem is that a lack on the behavioral level is taken to indicate a lack of psychological activity or subjective life or some underlying and fundamental deficit or diminishment in function. Recent research on the subjective experience of schizophrenia, however, suggests that observed behavioral lacks may not have direct equivalents or analogues on the experiential level (Sass 2000, 155). For example, flat affect is often accompanied by intense emotional reactivity at the experiential level despite lack of overt behaviors like affective expression, and asocial behavior can be accompanied by a stated yearning for contact with others. There are also positive-negative discrepancies on a single plane or level. For example, on the experiential level, Schneiderian first rank symptoms are considered positive symptoms because they involve the presence of experiences normally absent: hallucinations (e.g., hearing voices) and delusions (e.g., thought insertion). However, these symptoms imply the simultaneous absence of something normally present: a sense of ownership or intentional control (Sass 2000, 154). Moreover, the loss of some feature or function is often accompanied by an exaggeration of another in schizophrenia (Sass 2001).

These examples bring out that a deficit view of schizophrenia is too simplistic to capture the complex, unusual, sometimes even contrary qualities of the subjective experience of schizophrenia. The deficit view understands so-called negative symptoms "in purely quantitative or negative terms—as a simple privation of something normally present" (Sass 2001, 259) and emphasizes loss, absence, and decrease. Thus, it fails to appreciate the relevance of super-abundance or excess of "exaggerated cognitive capacities" (Sass 2001, 252) or to consider the possible advantages of alternative forms of functioning (Sass 2000-1a, 64) and the "qualitative uniqueness of schizophrenic orientations" (Sass 2001, 252). 16

Two lines of reasoning that underlie Sass's critique of the deficit view of schizophrenia can be used to develop the idea that a purely deficit approach impedes progress in our understanding of the phenomena themselves (and is thus objectionable in principle, at least for some phenomena). First, conceptualizing a phenomenon purely in terms of privation omits features, traits, capacities, and so forth that are, for some purposes and interests, better described as a higher degree of features, traits, capacities and so forth that are present to a lesser degree in typical individuals. Assuming that the deficit attributions are true, the kind of knowledge of the relevant phenomenon's quantitative variation of the normal that one gains from a purely privative explanation or description of that phenomenon is partial, at best. Such a construal leaves the phenomenon with no positive identity. It merely identifies what it is not, what it lacks, what it is missing by comparison with something else that has those attributes, capacities, and so forth.

Second, understanding a phenomenon purely through quantitative comparison with some other phenomenon is problematic (even if excesses and super-abundances are included in the comparisons) because it omits qualitative concepts when they are needed. That is, even if one conceives of the phenomenon as not just a matter of what is less or missing, but also as a matter of something more, a super-abundance, exaggeration, or excess, in some respects, one still neglects ways the phenomenon differs qualitatively. A purely quantitative descriptive and explanatory framework is a secondary, derivative, dependent, and reactive way to understand the phenomenon in question. In an important sense, the phenomenon has no identity of its own, independent from mere quantitative comparison with what is labeled normal.

The above observations about quantitative and qualitative concepts prompt a question about the relation between these two ways of conceptualizing variation. What effect, if any, does taking a purely deficit approach to understanding a phenomenon have on one's ability to conceive of that phenomenon in terms of qualitative difference? Feminist philosopher, Jay's (1981), analysis of the use of the A/Not-A dichotomy in understanding female-male gender differences suggest that, at least in some cases, a deficit approach is conceptually constraining in that it compromises one's ability to conceive of qualitatively different forms. Jay (1981) observes that an A/Not-A dichotomy underpins a purely deficit view. This kind of dichotomy construes A positively, in terms of presence, and not-A negatively, in terms of the absence of something A is or has. Phrased this way, the dichotomy prompts one to understand that which is labeled Not-A as a lesser form of A rather than as a form of its own. Only one term (A) has positive reality.

To return to the example of gender differences, Jay (1981) reminds us that historically woman and femaleness have regularly been characterized as Not-A. Woman has been understood as "misbegotten male" (Aristotle), "the lesser man" (Tennyson), or as a castrated man (Freud). Being a woman, on these conceptions, is the failure to become a man. Women are diminished or deformed men. Femaleness is a lack of maleness.

More generally, Jay's (1981) analysis suggests that relying on an A/Not-A dichotomy as a starting point and primary framework for understanding human differences, as a purely deficit view does, may have a harmful consequence for theorizing: it may inhibit creative, flexible conceptualization of those differences in our explanatory and descriptive frameworks exactly where we need it most—in the faithful depiction of qualitative variation. Reliance on this kind of dichotomy might make it more difficult to conceive of qualitatively different kinds of phenomena or encourage neglect of qualitative variation. One who understands a phenomenon in terms of an A/Not-A dichotomy finds it difficult "to conceive of the possibility of alternative forms of…order (third possibilities). Within such thinking, the only alternative to the one order is disorder" (Jay 1981, 54).

The lessons from Sass's critique of schizophrenia (2000, 2000-1a, 2000-1b, 2001) and Jay's (1981) analysis of the A/Not-A dichotomy in conceptualizing gender differences readily apply to the case of autism. Before discussing how the lessons apply, a few cautions about autism are in order. First, the kinds of differences associated with autism mentioned below are only a sampling of those reported. Second, we would do well to keep in mind the adage, common in autism communities, "If you have met one person with autism, you have met one person with autism". The heterogeneity of autism presents many challenges for nosology, and it is unclear how far one can generalize from particular personal accounts and research studies of differences associated with autism. Third, the impact of these differences on autists' development and everyday social functioning are not well understood.

Concerning the inclusion of quantitative excess, exaggeration, and super-abundance of traits, features, capacities and so forth, attributions of this kind are sometimes included in representations of autism. Take for example "hyperreactivity to sensory input" as a subcategory under "Restricted and repetitive behaviors" in the diagnostic criteria for autism spectrum disorder (DSM-5), the extreme male brain theory (Baron-Cohen and Hammer 1997, Baron-Cohen 2002; 2003), or the idea that autism can be thought of as the pathological excess of traits that are normally understood as desirable in their cultural context such as individuality, autonomy, and self-reliance in the Western world (Straus 2013). Nevertheless, when it comes to understanding autism as quantitative deviation from the normal, deficit attributions dominate. Autism, like schizophrenia and femaleness, has been characterized routinely as Not-A, as a deficit in a trait, feature, capacity and so forth that A is or has.

As for the inclusion of qualitative variation in explanations and descriptions of autism, there is a pressing need for more thoroughgoing reflection in scientific and philosophical investigations of autism on how autists' ways of being in the world, their subjectivity, experience, behavior, intelligence, and so forth differ qualitatively. Qualitative concepts are needed to capture the diversity and complexity of differences between autistic and typical individuals. Consider, for example, how Williams, an autistic adult, describes her perception of mood:

I could tell mood from a foot better than from a face. I could sense the slightest change in regular pace and intensity of movement of foot. I could sense any asymmetry in rhythm that indicated erraticness and unpredictability…Facial expression, by comparison, was so overlaid with stored expression, full of so many attempts to cover up or sway impression that the foot was much truer. I used sound in the same way, even breathing. Intonation aside, I could sense change in regular rhythm, pace, intensity and pitch. (quoted in Cole 1999, 96)

Here we have the description of a less ocular- and head-centric way of coming to understand a person's mood through non-verbal behavior than more typical ones (e.g., seeing emotions in facial expressions). But this is not a description of a point of view that is simply missing things. Nor is it a description merely of excess or exaggeration of what typical people do. What sense-modalities Williams relies on most, what she orients to and attends to, differ. Williams reports sensing the tempo and rhythm of breath, voice, and foot movement, qualities of kinesthetic behavior, as expressive of a person's mood.

Some autism rights activism calls for increased recognition of just this point. In some ways autism is better understood as a qualitatively different state of being with a qualitatively different form of order and intelligibility—not the typical order, but order nonetheless. In his autobiographical account, autistic adult, Stephen Shore (2003), suggests that, "the autism spectrum should be considered as 'another order' of being as opposed to a disordered, deviant way of existing" (v). Likewise, Amanda Melissa Baggs (2010), an autistic adult, writes, "This is about what is, not what is missing […] It is about the fact that those of us who are viewed purely as having had things taken away—as being essentially barren wastelands—are not shut out of the richness of life by being who we are. The richness we experience is not some cheap romanticized copy of the richness others experience" (n.p.). Baggs and Shore are calling for a conception of autistic personhood that has a positive identity of its own, where autistic personhood is not understood as a lesser form of typical individuals' personhood.

Another place where qualitative concepts are needed is in capturing differences in how autism manifests across individuals labeled autistic. 17 For example, sensory-perceptual differences are commonly reported in autism (Baranek et al. 2006; Bogdashina 2016; Grandin and Scariano 1996; Leekam et al. 2007; Williams 1996), but they manifest in a variety of ways. Some people with autism experience increased (hyper-) sensitivity to incoming stimuli while others experienced decreased (hypo-) sensitivity. The sensitivities occur across sensory modalities (i.e., vision, touch, taste, hearing, smell), are often idiosyncratic, can vary from hypo- to hyper-sensitivity within the same individual (Baranek et al. 2014), and may result in reacting differently to the same stimuli.

Qualitative variations in experience between individuals labeled autistic can also be seen in the sometimes-unusual ways they achieve intersubjective understanding. Compare Williams' description of coming to understand another's mood through non-verbal behavior from above with Tammet's (2006), another autistic adult's personal account of understanding others:

Numbers are my first language, one I often think and feel in. Emotions can be hard for me to understand or know how to react to, so I often use numbers to help me. If a friend says they feel sad or depressed, I picture myself sitting in the dark hollowness of number 6 to help me experience the same sort of feeling and understand it…By doing this, numbers actually help me get closer to understanding other people. (7)

While Williams describes how she uses non-verbal cues to help her understand others' moods, Tammet describes how his first language, numbers, helps him simulate, in his imagination, the feeling-in-context that he wishes to understand. Numbers are often associated with the impersonal, yet for Tammet they could not be more personal, more imbued with feeling.

These examples suggest that to capture the complexity and subtlety of autists' ways of being in the world, we need descriptive and explanatory frameworks for autism that take into account both the range of quantitative differences and qualitative differences, differences both in degree and kind from what is typical. As Murray (2013) succinctly puts it, autism "is somehow less, more, and something else" (53).