Ethnomethodological Studies of Visuality

The article considers how ethnomethodology (EM) studies visuality. Historically, there were four approaches to visuality in EM, those of visuality as observable activity, as images, as practices of vision, and as language constructions. Harold Garfinkel’s idea of witnessability implies that the phenomena of order exist in observable methods of their production. Observability, or witnessability, at this point can be an equivalent for visuality because it suggests an opportunity to be studied with visual methods. In such a case, any ethnomethodological study is a visual study because it implies a description of observable methods of order production. Besides this idea of observability, three separate projects of visual research were developed in EM from 1980s to the 1990s by Michel Lynch, Charles Goodwin, and Jeff Coulter. They all tried to present the practical approach to visual perception (in contrast with perception as an individual psychological process), but found solutions in studying different aspects of visuality, which were images, practices of seeing, and language constructions describing different modes of perception. This text considers the relationships between the three conceptions and Garfinkel’s initial idea of observability. It analyzes ideas which Lynch, Goodwin, and Coulter added to Garfinkel’s ethnomethodological program, and shows how other ethnomethodologists use these additions. The article demonstrates that together they produced EM’s approach to visuality, although none of these projects were completely realized inside EM. It is based on Garfinkel’s idea of observability, supplemented by the opportunity to study perception as a practical social achievement situated in local interactional contexts.

The idea of witnessability is key to ethnomethodology (EM). It implies that phenomena of order are produced through and in sequentially organized actions which are accountable for competent practitioners. Unlike the sociologies of "hidden order", 1 EM starts from the premise that social order exists in witnessably ordered practices rather than external social facts, such as rules or structures, which should be discovered to reveal sociality.
Such a conception presupposes that EM focuses on members' methods of social order production (ethnomethods) and recognizable details of practices. Despite the variety of fields and methods of study, 2 the principle of witnessability has directed EM since the 1960s (Korbut, 2013: 20-22) -although initially it wasn't specifically connected with visuality. 3 Witnessability of the phenomena of order is closely connected to being observable, and can thus be a synonym for visuality -as proposed by Rod Watson (Watson, 2005). 4 Understood in this way, visuality is a characteristic of practice rather than of a material object. Waiting for a bus or moving through pedestrian flow are visually organized practices, i.e. potential fields for visual analysis, which traditional sociology applies to material artefacts such as art objects or signs. In accordance with this position, any phenomena of order's description might be visual, but the visuality is not a separate field of study. Visuality is a characteristic that belongs to practice, and cannot be separated from practice.
If one equates visuality with witnessability, any EM study might be visual. It's therefore not clear what role other conceptions of visuality could play in EM. However, three more additional conceptions of visuality appeared in EM in the 1980-1990s. They were proposed by Michael Lynch in his studies of scientific representations, Charles Goodwin and his "professional vision" and Jeff Coulter's praxeology of perception. All of them aimed to demonstrate the practical basement of perception and argued against the idea of perception as a psychological process, but found solutions in different aspects of visuality: images, practices and language constructions describing "modes of perception".
In this article, I'm examining the relationships between four conceptions of visuality (visuality as witnessability, images, practices of seeing and descriptions) and their impact on contemporary visual studies in EM. I'll focus on the ideas that Lynch, Goodwin and Coulter added to wintessability, and trace how these ideas have been used in empirical studies close to EM. 2 For the review of the studies of 1970-1980s see for example : Garfinkel, 1986;Lynch et al., 1983. 3 Garfinkel doesn't equate witnessability to observability or visuality, although he often uses the terms interchangeably, as in his Studies in Ethnomethodology (1967). In the study of trust (1963) he concentrates on "perceived environments", and in his latter works, on audio-visual details (Garfinkel, 2002: 223, 283). At the same time, it seems that observability may be treated as one of the features of witnessability, which Goode demonstrates in his study of interaction with a deaf-blind child (Goode, 1990). For the child, the world is ordered not visually or observably, but witnessably, although an analyst discovers this orderliness though observations. 4 Before Watson, ethnomethodologists engaged in video-analysis used this understanding of witnessability without clarifying it (Goodwin, 1981;Heath, 1986;. I'll try to demonstrate that all the approaches have been integrated into EM's single conception of visuality. It's based on Garfinklel's idea of witnessability that is equalized with visuality, and implies the opportunity to study visual perception as a collective practical achievement, rather than an individual cognitive phenomenon.
However, the integration wasn't complete. The followers sustained Lynch, Goodwin and Coulter's studies of cognitive phenomena through observable practices -but used them as a secondary topic. The studies of visual perception haven't become a separate field in EM. The followers also didn't take most of the theoretical ideas that Lynch, Goodwin and Coulter contributed to Garfinkel, but borrowed the idea about the practical ground of perception.

VISUALITY AS IMAGES
Chronologically, Michael Lynch's studies of scientific representations were the first to add the analysis of images to the focus on the visibility of ethnomethods.
Lynch started to work on scientific representations during his laboratory studies along with Bruno Latour, Steeve Woulgar, Clauss Amann and Carin Knorr-Cetina (Amann & Knorr-Cetina, 1988;Knorr-Cetina, 1981;Latour, 1986;Latour & Woolgar, 1979;Lynch, 1985a). His interest was also formed by the STS (science and technologies studies) of that period -particularly the debates between realists and constructivists, and the question of whether scientific representations reflect reality or construct it. Instead of choosing sides, Lynch became interested in demonstrating how representations acquire their "realistic" form and how they relate to the scientific objects and the laboratory's social order (Lynch, 1991b: 207).
The analysis of scientific representations as material objects led Lynch to the clash with the conception of mental representations embedded in the cognitive theories of perception. That approach implied that representations are mental projections or retina's images (Lynch, 1988: 202), i.e. unobservable phenomena. By contrast, Lynch was interested in the ''externalized retina'' or material images, which could be "evidence of methodic practices, accomplished by researchers working together in groups, which transform previously hidden phenomena into visual displays for consensual 'seeing' and 'knowing'" (Lynch, 1988: 203).
The study of externalized retina implied that representations define what can be seen. According to Lynch, that's possible because representations contain the collective work on their creation. Lynch demonstrates that through the analysis of the image of cell in biology. The cell's model contains the work with a certain object which was observed (for example, a laboratory mouse), its processing with the lab equipment, standardization of the object's features, graphic-processing, highlighting one features and hiding the others. Other scientists perceive the resulting model naturalistically, and might use it as evidence to expose certain characteristics. Lynch considers several techniques of turning a scientific image into a 'natural' object (Lynch, 1985b(Lynch, , 1988 and demonstrates that laboratory practices embodied in the scientific images mediate visible fields and ways of perception. 5 In Lynch's 1980s work, images are "pre-linguistic modes of order production" that define what later appears in linguistic accounts (Lynch, 1985b: 52). 6 The idea that an image embodies the process of its creation and thereby forms perception, partly coincided with Garfinkel's program and partly resembled STS (particularly, Latour and Woolgar's inscription devices). Emphasizing the work embodied in images, Lynch followed the classical ethnomethodlogical study of scientific work (Garfinkel, Lynch, Livingston, 1981). However, his initial focus was on images as the results of generalized practices, forming the stable ways of perception. If it's possible to equate an image with instruction, that points how it should be read, then Lynch early studies remind an attempt to analyze practice based on the first part of the Lebenswelt pair only, i.e., on instructions rather than lived work. Thus, they go beyond classical EM, implying that image can presuppose practice (at least, the practice of perception).
Other authors mostly took from Lynch the first part, which corresponded with Garfinkel's program, and the interest towards scientific representations. Lucy Suchman's works on the representations in cognitive neuroscience described routine actions which create a scientific object instead of analyzing representations per se (Suchman, 1988). Dushan Bjelić in co-authorship with Lynch followed the same direction as Suchman studying experimentation (Bjelic, 1996;Bjelic, Lynch, 1992). They answered the question of how scientists make their objects visible, describing experimental technique and practice rather than the resulting images. Recent Phillipe Sormani's study on using a scanning tunnel microscope pursues the same kind of question. He pays much more attention to describe lived work of the lab and not the translation of data into representations or representations' properties. For Sormani, the question of how scientists see is a question of how the laboratory order becomes visible and accountable, which he answers through the classical ethnography, his own lab experience and video-analysis of the experiments (Sormani, 2014).
Lynch's studies about technologically mediated perception also influenced Goodwin's conception of visual practices (Goodwin, 1994: 601;Goodwin, Goodwin, 1996: 61). For Goodwin, technological tools and representations, as well as bodily practices, play a crucial role in structuring perception and making visual 5 Using the example from histology, he describes the process this way: "Graphic formats, instrumental fields and preparatory techniques in histology penetrate both the field of what is visible and the means for perceiving it. It is as though they operate as elements of an externalized retina, activating the perceptible and schematically processing it" (Lynch, 1985b: 59). 6 In another article on biological images, Lynch mentions that reading representations requires certain skills but doesn't elaborate that comment (Lynch, 1991). practices sustainable and reproducible (Goodwin, 2000a). However, Goodwin criticized Lynch's early works for treating representation as stable objects instead of paying attention to their use in local practices and tried to fill this gap in his own writings (Goodwin, 1996: 41). Similarly, the subsequent works of Lynch mostly focused not on the images but their usage (Lynch, Jordan, 2000).
In the same vein, the contemporary studies of science close to EM focus more on the use of images. They consider images more like fields for interaction rather than stable objects guiding perception. Morana Alač questions how scientists create representations in the interactional context, how the representations' form influence perception and how scientists make use of these representations in their work (Alac, 2011). Janet Vertesi uses ethnography to trace the process of creating Mars' pictures, from taking the decision on rover trajectory to images processing and their role in the following projects. She's simultaneously interested in the decision-making process, tusk distribution, and bodily practices of "seeing" pictures (Vertesi, 2015). These studies are more interested in local work than generalized methods of visualization: how neuroscientists do use fMRI pictures in their work, how Mars pictures are created, how physicists do work with STM in concrete situation here-and-now.

PRACTICES OF VISION
While the studies of scientific images related to the question of the perception only partially, that question was crucial for Goodwin' studies of "professional vision". In the 1990s Goodwin wrote a series of works about perception at work practices. At that time, his main task was to demonstrate the practical ground of vision and get away from the conceptions, that consider vision as an individual mental process. This direction of work grew out of Goodwin's earlier studies on the role of gaze in interaction (Goodwin, 1981), Lucy Suchman's project on the airport study at the Xerox PARC, and Goodwin's interest in Lave's and Rogoff's conceptions of social learning (Goodwin, Goodwin, 1996: 60 -63).
"Professional vision" -the main term from Goodwin which has become known far beyond EM -alludes to the first published article on that topic (Goodwin, 1994). 7 Here, Goodwin analyzes the videotape of the trial of policemen charged with beating Rodney King. Despite the presence of video-evidence, the jury initially acquitted the suspects because the defense witness convinced them that the policemen saw Rodney King's movements as aggression and attempts to attack (Goodwin, 2000a). Goodwin collates the court data with the data from archaeologic work where students are taught to classify soil by colors. He shows that as well as police officers, archeologists perceive the soil in a specific way, for example, see more color shades and use the difference in colors as data for their work. This ability to see and understand world selectively, in accordance with professional requirements, Goodwin calls professional vision (Goodwin, 1994: 606).
Of course, Goodwin wasn't the first to think about the selectivity of perception, though his interpreters outside EM noticed that idea more often than others. Goodwin's task was to demonstrate that professional vision is not a characteristic which belongs to a person and not an individual cognitive operation but a phenomenon which (in case of a court case) exists in the interaction between the defense witness, videotape of the beating, and the jury. The jury will see the signs of aggression only when a policeman through observable bodily gestures points at the scenes which must be seen, highlights the certain moments and aligns them with a "coding scheme", showing what the moments signalize. To "see" the color of soil, students should match its sample with Mansell scheme, react to the correction of more experienced archeologists, put the result in the table and mark it on the map. All of these implies that professional vision is produced through observable collective work: highlighting the episodes, using graphical representations, instructing and correcting. The vision is not an automatic process but a result of active bodily interaction with the world outside, structuring the world to highlight one features instead of others. The selectivity of vision is a collective practical achievement which is produced through observable and coordinated actions. In the article from 1994, Goodwin identifies three types of these actions: using coding schemes, highlighting, and using graphic representations (Goodwin, 1994: 606 -607). Goodwin demonstrates practical ground of vision in his studies on archaeological work (Goodwin, 2000b), oceanographic research vessel (Goodwin, 1995), an airport (Goodwin, Goodwin, 1996) and a chemical lab (Goodwin, 1997).
Demonstrating how artifacts and bodily practices structure perception, Goodwin didn't argue that vision is not a cognitive process. He argued against equating cognition with an individual brain (Goodwin, 1994(Goodwin, : 609, 1997Goodwin, Goodwin, 1996: 88). In sociology, this approach presupposes that there is an individual cognitive scheme behind vision, which is determined by social factors. Considering vision as an individual mental process, sociologists cannot study it directly. Before that, they should excavate the knowledge of vision with a method, e.g., through photo-interview (Harper, 2002), the survey on photographic tastes (Bourdieu et. al., 2014), or the analysis of visual artifacts which can be interpreted with a known interpretative scheme (Grasseni, 2007). For Goodwin, cognition and vision are phenomena situated in the interaction between people, tools, and artifacts (Goodwin, Goodwin, 1996: 88). Thus, they are observable and available for analysis. They are social due to their existence in the sequence of coordinated actions rather than due to the external social factors. In that, Goodwin's conception starts to resemble Edwin Hutchins' distributed cognition (Hutchins, 1995) 8 and differs significantly from the approaches which treat vision as a cognitive operation presupposed by external social structures. Therefore, the gap between the works of Goodwin and, for example, Pitirim Sorokin, who explains the differences in "vision" through knowledge structures (Abramov, 2016: 303), is more significant than it might seem to be.
Goodwin opposes the idea that external structures, like language, define vision. That position is best demonstrated in his study of experimentation in a chemical laboratory (Goodwin, 1997). Empirically, the article considers an applied experimental task -identifying that an indicator, which signalizes the end of the experiment, has become black. The task is difficult for the experimenters because they have to define every time whether the color is deep enough to call it black. Theoretically, the paper argues with the structural approach to color perception in linguistics and cognitive anthropology. For the linguists who followed Saussure, the structures of perception are language structures (that was named Shapiro-Whorf hypothesis). Developing this idea, Berlin and Kay (1969) showed that the systems of classification in all languages follow the same principles; and that similarity might be connected with the structure of human visual system. Goodwin argues against that position with the idea of color situated in practice. He demonstrates that the category of "black" appears to be useless in the experiment. To conduct the experiment successfully, the chemists should define their own criteria of "black" relevant for the task. Thus, they create new categories to define the color (an indicator of the appropriate black is called "gorilla fur"). Demonstrating how the chemists cope with the task, Goodwin shows that color perception is situated into the local work and specific practice -a chemical experiment in that case. The experiment demonstrates that there is no a color "in itself", it exists only as a part of practice.
Goodwin's studies of visual practices have become known far outside EM, though sometimes highly revisited or through generalized citations. 9 Some of the ethnomethodolohists also misunderstood his works. Critically reviewing Goodwins' works, Rawls gratified him for popularizing well-known idea that professionals see the world differently and can see the objects which the others can't. Citing Ethnomethodology's Program (Garfinkel, 2002), she claims that for Garfinkel every interaction with a material object presupposed socially-defined vision (Rawls, 2009: 713), so it might be concluded that Goodwin's studies in that direction were redundant at some sense.
In fact, Goodwin never claimed that the selectivity and practical ground of perception belongs to professional environments only. In the first article on that topic, he states that the practices of coding or highlighting are omnipresent, though professional settings illustrate the social basements of vision better than others (Goodwin, 1994: 630). Likewise, though Garfinkel didn't concern individual psychological processes in his studies, he also didn't criticize individualism precisely. In that sense, Goodwin's practical studies of psychological processes sustained Garfinkel's logic but also continued it. Perhaps, the practical conception of vision could be built on Garfinkel's program solely, but Goodwin was also drawing on Vygotsky's activity theory, cognitive anthropology of Lave, Wangler and Rogoff, and Hutchins' theory of distributed cognition. Eventually, the studies of vision led Goodwin to his own theory of action (Goodwin, 2000c), where he abandons, along with Hutchins, the idea of individual agent and concentrates on the role of semiotic resources in building the sociality (Streeck, 2015).
Judging by the number of citations, at some point mentioning Goodwin has turned to be almost obligatory in EM's visual studies and beyond. However, as in the case of Lynch, ethnomethodologists rarely used those ideas of Goodwin which went beyond Garfinkel's program. Goodwin almost didn't have followers who would continue to work with the same theoretical problems and resources. Almost nobody has concentrated on the nature of perception and cognition -the topics which were central to Goodwin. His ideas about the social ground of perception and practical expression of cognitive phenomena were supported, the conception of embodied participation framework has been used to a much smaller extent.
Ethnomethodologists use it to focus on interaction and nonverbal resources, but stay adherent to the "'humanist' conception of agency" (Streeck, 2015: 432) with an individual agent who can, for example, use semiotic resources for self-expression. Quite in contrast, for Goodwin, the agency is blurred in the interaction between bodies, gestures and semiotic objects.
The researchers who focused on "observable", "visible" or "professional vision", following Goodwin, were much more involved in the description of concrete practices rather than the discussion of theoretical problems concerned by Goodwin. In that sense, the studies of how building working groups becomes visible (Kawatoko, Ueno, 2003), using video in medical work (Mondada, 2003), and the numerous studies of instructions (Alby, Zucchermagilo, 2008;Gåfvels, 2016;Lindwall, Ekstorm, 2012, Rystedt et. al., 2011 follow Goodwin only in part. They use Goodwin's ideas to include cognitive processes in the analysis of practices, but keep concentrating on the details of practices. The works of Aug Nishizaka might be among the exceptions since, as well as Goodwin, he focused on the nature of vision in the 2000s (Nishizaka, 2000a(Nishizaka, , 2000b. He changed that emphasis later but remained close to Goodwin with his attention to the practices of structuring perception (Nishizaka, 2006(Nishizaka, , 2014. Morana Alač has also used Goodwin's ideas on semiotic resources more actively than others (Alac, 2011), though combining them with the STS approach to scientific representations and the ideas of Umberto Eco and Charles Pierce.

PRAXEOLOGY OF PERCEPTION
Jeff Coulter, Ed Parsons, and Wes Sharrock elaborated the third conception which influenced EM studies and could have become an alternative to witnessability (Coulter, Parsons, 1990;Sharrock, Coulter, 1998, 2003. As well as Goodwin, Coulter with the coauthors aimed to abandon understanding of vision as an individual cognitive operation and consider it as a situated practical achievement (Coulter, Parsons, 1990: 251 -252). The basic difference between the approaches is that Coulter also tried to abandon approaching vision as a homogeneous process. Drawing on philosophy of language, Coulter proposes that a study of visual should focus on language constructions describing the ways of visual orientation or different modalities of perception: observing, noticing, searching, gazing, etc. (Ibid.: 260-261). One should focus on the usage of these "visual" verbs and the contexts of their usage instead of studying visual perception in general or focusing on one of the well-studied modalities like "seeing".
In contrast with Lynch and Goodwin's studies, the praxeology of perception emerges out of theoretical polemic rather than empirical studies. Among the disputes parties were Jerry Fodor and Zenon Pylyshyn -the philosophers oriented to cognitive science -and James Gibson, a psychologist.
Fodor and Pylyshyn represented the 'cognitive -constructivist' approach to perception which was the main target of Coulter and Parsons' criticism (Coulter, Parsons, 1990: 255). The approach presupposes that visual perception consists of two parts: the physical processing of information by neurons and the translation of information into concepts. The interpretation, translation of information into mental representations, or concepts, is a social process. The other part is a cognitive operation, which should be studied by psychology. Coulter criticized this division and tried to abandon the psychological part, saving the "concept-boundedness" of vision, i.e. the tie between vision and language (Coulter, Parsons, 1990: 255 -256).
At the first stages of the dispute, Coulter allies with Gibson and his ecological theory of perception. Gibson proposes the theory of direct perception without mental representations as mediators. He claims that structuring of information doesn't occur through cognitive operations -the information flow is already structured by the affordances of the outer world. Coulter also supports Gibson's criticism of the studies of static perception. According to Gibson (and Coulter), perception is always embedded in other activities, and it should be studied in the contexts of its implementation.
Though Coulter and Parsons support the intention to contextualize perception, they claim that Gibson withdraws the next necessary step -to stop treating visual perception as a homogeneous phenomenon. If perception is always situated, thus, instead of studying it "in general", one should pay attention to the different modalities of perception. Only a few of these modalities have already been studied, including: "seeing", 10 "looking", and "scanning". The next step is to broaden the list and analyze the constructions, which we use to discuss perception in concrete situations. 11 Coulter and Parsons criticize EM studies of perception on the same grounds. Discussing Goodwin's studies of gaze in conversations, Coulter states that Goodwin ignores the ordinary meanings which the words as "gaze" and "gazing" have in conversations. Saying that somebody "gazes" at the other while talking signifies that she is not engaged in the conversation completely. Saying that somebody gazes at the passer-by means too intensive attention. Using the words like "gaze" as generalizing concepts, the researcher can't grasp these meanings (Coulter, Parsons, 1990: 265 -266). Coulter appeals to the same criticism, discussing "seeing" and "viewing" in Harvey Sacks' membership categorization analysis (MCA). In one of the first papers on MCA, Sacks demonstrated that perception is category-bound: one tends to see the participants of certain activities as members of the category related to that types of action (Sacks, 1972: 338 -339). He demonstrated that analyzing the famous fragment "The baby cried. The mummy picked it up". When an observer sees such sequence of actions, she sees that a mother picks up a baby rather than a woman picks up a man. Sacks calls this mechanism the viewer's maxim. However, the example is not completely correct for Coulter, since viewing can be complemented by such objects as TV-programs, pictures, or parades, while the described scene could be "noticed", "observed" or "witnessed" (Coulter, Parsons, 1990: 264). The alternative research program may be analyzing perception in the categories embedded in the settings of its usage.
As an example of that research program, Coulter and Parsons trace the use of "noticing", which is much more restricted compare to "seeing". They provide a general overview of the situations where perception can be described as noticing and distinguish them from other perceptual modalities, including "noticing that" or "not noticing" (Coulter, Parsons, 1990: 266 -269). However, despite the illustration, the empirical program of praxeology of perception remains vague. For example, it's not clear whether it should be based on "lived work", members' accounts or typical use of language. It also remains unclear, whether Coulter and his 10 Further on, citing Ryle, Coulter and Sharrock argued that "seeing" or "perceiving" can't be used for studying psychological process because they are words for achievement rather than process (Sharrock, Coulter, 1998: 149-150;2003: 77). 11 In the subsequent publications Coulter extends his critique of Gibson's theory, after having discovered remnants of cognitivism in it. In co-authorship with Sharrock, he later criticizes the notions of affordance and information. Since affordances are situated, there is no sense to focus on them; one should focus on the situations where affordances acquire their meanings. There is also no point in adding new terms like affordance or information, since the researchers should keep adherent to the concepts which members use for "seeing" (Sharrock, Coulter, 1998: 162). coauthors are interested in language exclusively or bodily practices can be added to the analysis.
In the following works, Coulter and coauthors focus on the theoretical discussions with Gibson, Fodor, and Chomsky (Sharrock, Coulter, 1998, 2003, but do not elaborate their empirical program. That might be a reason why praxeology of perception has not been completely implemented in other research projects. Empirical studies either develop the point on the practical ground of perception, which is shared by all EM approaches to visuality (Alac, 2011;Busher et. al., 2000;Evans, Fitzgerald, 2016;Heinemann, 2016;Koschmann et. al., 2010), or borrow selected concepts from the program, focusing on separate practices rather than perception itself Licoppe, 2017;Llewellyn, Burrow, 2008;Sormani, 2014).

VISUAL AND OBSERVABLE IN CURRENT EM STUDIES
The approaches presented in the works of Lynch, Goodwin and Coulter, treated visual as a separate field of studies. That line of study was hardly supported in the subsequent works. That doesn't mean that ethnomethodologists stopped visual studies, but the questions about the nature of vision were replaced by the interest to the specificity of practices.
In contemporary EM studies, "visuality" refers to the method (video analysis) and to the focus on observable methods of organizing practice. Very roughly, contemporary studies of visuality in EM can be divided into several overlapping thematic fields. They are: studies of workplace (Luff, Hindmarsh, Heath, 2000), nonverbal interaction (Heinemann, 2016;Kidwell, 2015;Mondada, 2016;Nishizaka, 2013); scientific work and representations in science (Alac, 2008(Alac, , 2011(Alac, , 2014Sormani, 2014); instructions and learning (Bernhald et. al., 2007;Gåfvels, 2016;Lindwall, Ekstrom, 2012;Lindwall, Lymer, 2008;Nishizaka, 2014); interaction with techologies and computer interfaces (Alby, Zucchermaglio, 2008;Heath, vom Lehn, 2008;. These studies sometimes borrow topics from Lynch, Coulter and Goodwin (images, professional perception, language), but concentrate on the description of practices, i.e., more likely follow the initial Garfinkels' interest rather than any of the particular programs of visual studies. Vision/perception and the ways of its implementation arise as separate topics only to the extent that they help to describe the specificity of practices and local interaction.
Four approaches to visuality, which historically existed in EM (visuality as witnessability, images, practices of vision and its descriptions), might have become competing, but empirical studies integrated them into the uniform conception of visuality in EM. This conception is based on Garfinkel's idea on witnessability/visuality as practice' property, and implies that visual perception can be studied as a joint practical achievement rather than an individual mental process.
In summary, on the one level one could diagnose the failure of Goodwin, Coulter and Lynch's projects, since they have never been completely implemented in EM or outside. The followers took some of the topics which were interested for the authors, but save only the ideas corresponding with Garfinkel's position and the idea of witnessability. On the other hand, the starting points of these conceptions formed EM approach to visual perception and other cognitive phenomena as collective practical achievements situated in practice and interaction. They added the possibility to study cognitive processes to Garfinkel's conception, but get over cognitivism as a principle of explanation.