[OPR] Luginbühl & Schneider: Medial Shaping from the Outset

On this page you can download the discussion paper that was submitted for publication in the Journal for Media Linguistics. The blogstract summarises the submission in a comprehensible manner. You can comment on the discussion paper and the blogstract below this post. Please use your real name for this purpose. For detailed comments on the discussion paper please refer to the line numbering of the PDF.

Discussion Paper (PDF)

Blogstract of

Medial Shaping from the Outset. On the Mediality of the Second Presidential Debate, 2016

by Martin Luginbühl & Jan Georg Schneider

In the present article we argue in a first part that all communication is medial in the sense that every human sign-based interaction is shaped by medial aspects from the outset (and not just as a secondary effect), and we propose a dynamic, semiotic concept of media that focuses on the process-related aspect of mediality and defines media as social procedures of sign processing (cf. Schneider 2017). The traditional reification of media still has strong impact not only in German linguistics and communication science, but also in the international discourse on media. We criticise the reification of media by arguing that all media are technical media, but the technical aspect cannot be reduced to materiality. Our dynamic concept takes into account the narrow link between “sign” and “medium” in social interaction and is therefore relevant as a theoretical and methodological basis of multimodal interaction analyses.

In the second part, we test the applicability of the proposed definition using as an example the second presidential debate between Hillary Clinton and Donald Trump in 2016, which shows in detail how the sign processing during the debate is continuously shaped by structural aspects of television and specific traits of political communication in television (cf. Luginbühl 2019). These structural aspects – the technical infrastructure, para-interaction, entertainment, the fourth estate, political propaganda and so on – can potentially conflict with each other, which leads to, and is exploited by, specific practices on the part of the hosts, the politicians and the studio audience. The way oral communication is processed is therefore contoured from the start by the whole medial procedure, including the ways in which turn-taking is organized, topics are introduced and avoided, face work is done and controversies are cheered on or ended, and where people move or look, i.e. how the spatial arrangement and camerawork create meaning and how the protagonists both use the affordances of this special mediality.

Therefore, it is not adequate to separate the technical aspects of the medium, the “hardware”, from the processual aspects and the structural conditions of communication. These three aspects together constitute the mediality of a medium, i.e. of a medial procedure. What German linguists call “communication form” is included in the medial procedure. If we separate these aspects from each other, it is impossible to adequately analyze the “medial traces” (cf. Krämer 1998) they leave behind. Brock and Schildhauer’s (2017) definition of communication form avoids separating these aspects by integrating the concept of medium into it. As we argue, however, the concept of communication form can be dispensed with altogether if we begin from a holistic understanding of media and then describe the specific medium in question in its specific granularity. Overall, the analysis demonstrates that, even in this staged situation, face-to-face communication must already be regarded as an inescapable medium of human communication and has a mediality from the outset.

References

Brock, Alexander / Schildhauer, Peter (2017): Communication Form: A Concept Revisited. In: Brock, Alexander / Schildhauer, Peter (Hg.):  Communication Forms and Communicative Practices. New Perspectives on Communication Forms, Affordances and What Users Make of Them. Bern/Berlin: Peter Lang, 13-43.

Krämer, Sybille (1998): Das Medium als Spur und als Apparat. In: Krämer, Sybille (Hg.), Medien, Computer, Realität. Wirklichkeitsvorstellungen und Neue Medien. Frankfurt am Main: Suhrkamp, 73-94.

Luginbühl, Martin (2019): Mediale Durchformung: Fernsehinteraktion und Fernsehmündlichkeit in Gesprächen im Fernsehen. In: Marx, Konstanze / Schmidt, Axel (Hg.): Interaktion und Medien. Interaktionsanalytische Zugänge zu medienvermittelter Kommunikation. Heidelberg: Winter (= ORALINGUA, Bd. 17), 125-146.

Schneider, Jan Georg (2017): Medien als Verfahren der Zeichenprozessierung. Grundsätzliche Über­­legungen zum Medienbegriff und ihre Relevanz für die Gesprächsforschung. In: Gesprächs­for­schung – Online-Zeitschrift zur verbalen Interaktion 18 (2017), 34-55. Online unter: http://www.gespraechsforschung-online.de/fileadmin/dateien/heft2017/ga-schneider.pdf

4 Replies to “[OPR] Luginbühl & Schneider: Medial Shaping from the Outset”

  1. RedaktionOctober 4, 2020 at 10:57Reply

    GUTACHTEN VON WERNER HOLLY

    Gutachten zu Luginbühl/Schneider: “Medial Shaping from the Outset.”

    Man kann der Titel-These des Beitrags uneingeschränkt zustimmen. Was sonst könnte Medienlinguistik tun, als die grundsätzliche Medialität von Kommunikation aus einer linguistischen und zugleich kulturwissenschaftlichen Perspektive aufzuzeigen.

    Die Autoren analysieren eine wichtige und auch wieder aktuelle US-presidential debate (Clinton/Trump) und thematisieren dabei Aspekte, die seit den 80er Jahren gängig sind: „the technical infrastructure, para-interaction, entertainment, the fourth estate, political propaganda and so on“ (1059-1061); sie betrachten auch turn-taking, Themenbehandlung, face-work, Körpersprachliches.

    Sie schicken ihrer Analyse eine mehr als 11-seitige theoretische Diskussion voraus, die allerdings einiges Fragwürdige enthält. Hier ist auf drei Bereiche hinzuweisen:

    1. Zunächst zum (hier zentralen) Medienbegriff, der angesichts der eigenen Warnung Widersprüche enthält und zu kategorialen Problemen führt. Die Zeilen 257-260 formulieren die Warnung:

    „The classic ‚Socratic‘ question ‚What is a medium?‘ promotes the reification of media and leads to categorization problems that cannot be solved convincingly.”

    Trotz dieser Einsicht werden aber mehrfach Definitionen mit Antworten auf die “What is?”-Frage gegeben:

    „Media are (kursiv-Hervorhebung von W.H.) social procedures of sign processing.” (8-9)

    “We understand media as (kursiv-Hervorhebung von W.H.) socially constituted procedures of sign processing.” (231-232)

    “Historically, the expression medium (kursiv im Text) has always referred to mediation, materiality, and potentiality; it is (kursiv von W.H.) a socially constituted procedure of sign processing.” (388-391)

    Danach wird eine Zitiergröße bemüht: „As Christian Stetter (2005, 91) puts it, a medium is a symbolizing procedure operating over a substrate or conglomerate of apparatuses.“ So hat das Stetter allerdings nicht gesagt. Die Stelle bei Stetter (2005, 91), die nicht explizit zitiert wird, lautet (in einem Derrida-Kontext): „Auch ein solches Spur-setzen ist nur in einem Medium möglich, in einem über einer Apparatur welcher Art auch immer operierenden Verfahren.“ Die von Schneider unterstellte Gleichsetzung von Medium und Verfahren (Zeile 274 „medial process (=medium)“) ist bei Stetter so klar nicht, wie von Schneider formuliert. Deshalb wird eine andere Form der Autorisierung versucht, indem eine wörtliche Formulierung von Stetter nachgeschoben und jetzt als bloße Formulierungsvariante interpretiert wird: „Viewed from the other side, one could also say that …“: dann folgt (in Übersetzung) als wörtliches Stetter-Zitat: „Ein Medium […] ist eine in Operation gesetzte Apparatur, sodaß durch diese Operation etwas, nämlich eine Darstellung von bestimmter Gestalt hervorgebracht wird.“(Stetter 2005, 74) Unter der Hand ist aus einer Apparatur ein Verfahren geworden; so auch schon in Schneiders Habilschrift von 2008.

    Mit diesem – nennen wir es – „Zitierkunststückchen“ wird der Eindruck erweckt, die definitorische Gleichsetzung von Medium und Prozess, welche die Autoren hier zu einer markigen These machen wollen, sei mit den Weihen derjenigen versehen, welche die Operativität von Medien (als eine Eigenschaft, nicht als Essenz) herausgearbeitet haben (Stetter, Jäger u. a.). Zieht man dagegen deren eigene Formulierungen heran, findet man ein sehr viel vorsichtigeres Vorgehen. Der kühnste Schritt von Stetter selbst in Richtung auf Prozesshaftes ist nämlich ein Satz mit zwei ‚Hecken‘ (in diesem Sinne, verkürzt gesprochen) und einem zusätzlichen Klärungsversuch (genauer gesagt):

    „Medien in diesem Sinne sind, verkürzt gesprochen, symbolisierende Performanzen, genauer gesagt: das, was an der performance reiner Vollzug ist.“ (Stetter 2005, 74)

    Unmissverständlich und kategorial klar formuliert Jäger in seinem Handbuchartikel zu ‚Medialität‘ (Jäger 2014), was den Witz einer Betonung des Operativen oder Prozeduralen der Medien ausmacht. Mit dem Begriff der Medialität – so Jäger – „tritt also neben die Analyse der verschiedenen Medien und ihrer kommunikativen und kognitiven Leistungen die Reflexion des intensionalen Bestimmungsmomentes, das das extensionale Feld des Medialen zusammenhält: die Reflexion der Medialität der Medien.“ (Jäger 2014, 107). Medialität ist für ihn ein „Suchbegriff“, der den „Blick auf Eigenschaften des Medialen“ lenkt (ebd.). Hier wird der theoretische und begriffliche Ort des Prozeduralen ungleich umsichtiger und präziser formuliert:

    „Aus der Perspektive der Medialität tritt die Operativität der Medien als eine ihrer grundlegenden Eigenschaften hervor.“ (Jäger 2014, 110)

    Medien haben also die Eigenschaft, operativ zu sein, sie prozedieren etwas, aber sie sind nicht selbst Prozeduren. Eine angemessene Formulierung wie „Prozeduren in Medien“ zeigt, wie schräg und tautologisch ein Medienbegriff wird, der Medien mit Prozeduren einfach identifiziert statt ihnen Prozeduren zu attribuieren. Die Kurzversion in Zeile 274 „medial process (=medium)“ stiftet eher Verwirrung.

    Natürlich ist es sinnvoll und manchmal vielleicht auch noch notwendig, gegen die Reduktion des Medialen auf eine bloße Transportfunktion zu argumentieren, was viele seit vielen Jahren getan haben. Auch hier formuliert Jäger (2014, 110) unmissverständlich: „Das Mediale der Medien zeigt sich durch die spezifische Art und Weise, in der sie prozedieren: nämlich als Figuren der Vermittlung (Krämer 2003, 79; ebenso Sandbothe 2003), in deren Vollzug die Elemente dessen, was vermittelt wird, allererst konstituiert werden.“

    Wenn man der Verflüssigung des Medienbegriffs durch Betonung des Operativen, Prozeduralen dienen will, wird dies eher durch eine Kippfigur möglich sein, wie sie mit der Langue-/parole-Terminologie des „Archimediums Sprache“ musterhaft und beispielgebend bereitliegt. Überhaupt scheinen solche Kippfigur-Begriffe der Dialektik von Struktur und Prozess, die in vielen Bereichen am Werk ist, gegenüber einseitigen Radikalisierungen eher gerecht zu werden.

    Dies alles möchte man den Autoren zu denken geben, wenn man sie auffordert, im Dienste der kategorialen Klarheit und eigenen Widerspruchsfreiheit der Verlockung allzu starker Thesen zu widerstehen. Es wird semantisch schräg, wenn jetzt z.B. alle bei Jäger (2014, 108) genannten „anthropologischen Artikulationsmedien“, Stimme, Hände und Gesicht, Vorgänge sein sollen.

    2. Institutioneller Aspekt (Zeilen 91-189): hier ist die Rede von „recent media theories“ (91) und „new institutionalism“ (113f.). Alles, was dann vorgebracht wird, ist allerdings keineswegs neu. Von spätestens den 70er Jahren bis heute haben angelsächsische Arbeiten, insbesondere die britischen „Cultural Studies“, aber natürlich auch deutsche Arbeiten zu Medien (Massenmedien, Individualmedien) Institutionelles thematisiert; besonders die deutsche Kommunikations- und Medienwissenschaft war immer durch und durch politisch; sie hat sich eher zu viel als zu wenig um Institutionelles gekümmert; so auch selbstverständlich die (relativ wenigen) Sprachwissenschaftler, die sich schon früh mit Medien beschäftigt haben (z.B. Straßner, Dieckmann und nicht zuletzt Harald Burger, natürlich viele andere mehr). Wenn es überhaupt blinde Flecken gegeben hat, dann auf Seiten der Kommunikations- und Medienwissenschaft beim Thema Sprache, wie einige auch offen eingestanden haben. Richtig ist also auch Jägers These von der Sprachvergessenheit der Medientheorie; was die ebenfalls von Jäger beklagte Medienvergessenheit der Sprachtheorie angeht, bezieht sie sich eben auf die Sprachtheorie, nicht auf die Linguisten, die sich als Teil der Angewandten Sprachwissenschaft sehr wohl und von Anfang an im Zusammenhang mit Medien explizit und mit großem Interesse um Institutionelles und Politisches gekümmert haben.

    Nur kleine Hinweise aus eigener Perspektive zum hier einschlägigen Thema Fernsehen: Holly/Kühn/Püschel (1986) „Politische Fernsehdiskussionen“, Kapitel 3 „Institutionelle und mediale Bedingungen von Fernsehdiskussionen“. Das schmale Bändchen Holly (2004) mit dem Titel „Fernsehen“: 1.3 „Fernsehen als öffentlich institutionalisiertes Massenmedium“. Usw., usw.

    Wenn Silverblatt (2004) neu entdeckt haben soll, dass Massenmedien soziale Institutionen sind, dann muss man z.B. erinnern an: Meyrowitz (1985), Fiske (1987), die „Social Semiotics“-Arbeiten, die ebenso lange hochpolitisch und soziologisch gut informiert mediale Kommunikation beschrieben haben. Wenn Shoemaker und Reese (2014) ernsthaft behaupten wollen, dass Journalismus als unabhängige, objektive Autorität wahrgenommen wurde, dann muss man erinnern an die Diskussion um hegemoniale Lesarten, um Rassismus in Medien, um die ganze Stuart Hall-Schule; in Deutschland gab es reichlich Lärm um „Rotfunk“, „Schweigespirale“ oder „Boulevardisierung“ und wie die Schlagwörter alle hießen, die seit den 70er-Jahren unüberhörbar waren.

    Zeile 143: „According to new institutionalism, however, they [journalism and social institutions] cannot be separated at all: mass media have never represented reality ‚objectively‘, but are themselves ‚political actors‘.“ Wer hätte das gedacht? Es gab vor vielen Jahren schon mehr als diese Binsenwahrheit, sogar sehr präzise und elegante Analysen: Man denke nur an Emanuel Schegloffs berühmten Aufsatz von 1988/89 zum „Rather/Bush-Encounter“, berühmt, weil er rein sequenzanalytisch die Stelle aufzeigt, an der der prominente TV-Journalist Dan Rather vom Interview-Format abweicht, indem er durch einen dritten Schritt Bushs Antwort qualifiziert und damit aufgibt, durch Display von ‚neutrality‘ die Gesprächsführung in alleiniger Regie beanspruchen zu dürfen. Confrontainment, wie es sich zeitweilig durchgesetzt hat, war dann wirklich nicht mehr als objektiv zu verstehen. Wir sprechen von den 80er Jahren.

    Die „4. Gewalt“ muss man nicht „neu“ entdecken. Insgesamt sollte statt Neuigkeitsanspruch eher ein Überblick über oder wenigstens ein Verweis auf die lange Tradition des Zusammenhangs von Sprache, Medien generell, Institutionen und Politik stehen. Kleiner Hinweis: Allein schon die Sprachgeschichte von Peter von Polenz enthält nützliche Zusammenfassungen, von der frühbürgerlichen Zeit an aufwärts.

    3. Kommunikationsformen. Ich empfehle den Abschnitt 1.3 („Medium and Communication Form“) wegzulassen. Er enthält zahlreiche Fehler und fällt hinter den aktuellen Forschungsstand (s. dazu Meiler 2018, 122ff.) weit zurück. Trotz des nützlichen Überblicks bei Brock/Schildhauer (2017) ist auch dort das eigentliche Leistungspotenzial des Begriffs nicht verstanden. Das zeigt sich vor allem an der Vorstellung einer Art von Kontinuum zwischen den Begriffen „Medium, Kommunikationsform, Genre“: „Where does medium end and communication form start, and where does genre take over“ (Brock/Schildhauer 2017, 18). Völlig unverständlich erscheint mir in diesem Zusammenhang ein Satz, der deren „currently most sophisticated concept“ erläutern soll: „they try to separate the mediality of communication from conventional practices/genres while maintaining a distinction between medium and communication form:“ (Zeilen 333-336). Hier scheint mir vor allem wichtig, einen eingeschränkten Praktikenbegriff zu überwinden, der offensichtlich zu Missverständnissen führt; außerdem wird man wohl ohne Vorstellungen von Hybridität nicht sehen, worin der Witz der Kommunikationsformen bestehen kann.

    Dass der in Holly (2011, 155) verwendete Praktiken-Begriff in Brock/Schildhauer (2017, 14) nicht verstanden wird, zeigt sich schon darin, dass er im Widerspruch („at odds“) zum Begriff strukturellen Aspekten gesehen wird, von denen ebenfalls die Rede ist, und zwar „von Strukturen, die als Bedingungen der Möglichkeit von Kommunikation an der Genese von Sinn beteiligt sind. Auch sie formen als Sets struktureller Arrangements Sinnpotenziale, indem sie als Dispositive den Gebrauch von Zeichen prägen.“ (Holly 2011, 150) Hier muss man darauf hinweisen, dass in vielen soziologischen Zusammenhängen auf die Dialektik von Strukturen und Handlungen hingewiesen wird, von Berger/Luckmann (1966) bis Giddens; seine Theorie der Strukturation zeigt sehr schön, „diesen Prozeß der Selbstformung von Handeln und Strukturen zugleich, in dem nicht kurzerhand präexistente Strukturen Handeln hervorbringen – und damit soziale Reproduktion sichern – sondern im temporalen Handlungsvollzug auch die Regelstrukturen potentiell modifiziert werden.“ (Reckwitz 2007, 321). Dazu kommt, dass ein Begriff wie Praktiken ein „festere“ und eine „flüssigere“ Lesart hat, ähnlich wie die Unterscheidung von ‚Handlung‘ und ‚Handeln‘ bei Schütz und die Unterscheidung von ‚Handlungsmuster‘ und ‚Handlung‘ in der Pragmatik. Diese dialektische Sicht reicht bis in aktuelle praxeologische Überlegungen. So trägt ein solcher Terminus – richtig verstanden – dazu bei, das Dilemma von einseitig konstruierten Sozialbegriffen aufzuheben, das deterministische Vorstellungen oder der Verzicht auf jegliche Institutionalisierung für die Erklärung sozialer Strukturen und ihrer Dynamik mit sich bringen.

    Man kann sogar so weit gehen, in dieser Kippfigur von Struktur und Handeln im Falle der Kommunikationsformen dem Form- und damit Strukturaspekt den Vorrang zu geben wie Meiler (2018). Bei Domke (2017) wird eine type/token-Unterscheidung vorgeschlagen, indem „Rahmen“ von Kommunikationsformen von aktuellen Realisierungen unterschieden werden; auch Pflaeging (2017, 171, Anm. 4) unterscheidet „patterns“ und „realizations“ (wenn auch unnötig kompliziert).

    Absurd erscheint mir hingegen, den Kommunikationsformen kulturelle Prägungen absprechen zu wollen, als ob nur Genres kulturell seien (so wie content oder cultural practices), nicht aber die Kommunikationsformen (s.o.). Natürlich haben auch ‚Formen‘ kulturelle Prägungen. Ein rasches Ende finden solche Engführungen in der Akteur-Netzwerk-Theorie von Latour, wo solche Scheidungen endgültig aufgegeben werden und auf die hybriden Strukturen hingewiesen wird, die Menschen und Dinge immer schon verbunden haben.

    Wenn die Autoren zu Recht begrüßen, dass die Trennung von Medium und Kommunikationsform bei Brock/Schildhauer (2017) endlich überwunden werde, dann kann man sie darauf hinweisen, dass sie dazu bei Holly (2011, 155) schon lesen können: „Kommunikationsformen sollen hier als ‚medial bedingte kulturelle Praktiken‘ verstanden werden.“ Und weiter: „Der strukturelle Ort, wo beide elementaren Prägekräfte der Kommunikation, Medialität und Kulturalität, sich verbinden, sind die Kommunikationsformen. Sie sind demnach die medial, historisch und kulturell verankerten kommunikativen Dispositive, die sich auf der Basis verfügbarer technischer Möglichkeiten und sozialer Bedürfnisse allmählich herausbilden und weiterentwickeln […].“ (ebd.)

    Warum begnügt man sich nicht mit dem Medienbegriff? Die Autoren weisen zurecht darauf hin, dass es sich hierbei um ein „zoom concept“ handle, dazu – wie man weiß – von beträchtlicher Reichweite und Unübersichtlichkeit. In der Analyse konkreter empirisch zu beschreibender Kommunikationen ist es aber sehr hilfreich, wenn die Vielfalt der medialen Strukturen und sich überlagernden (zumal technisch verfügbaren) Medialitäten jeweils situationsspezifisch erfasst wird, wie dies schon in einem heutigen komplexen Fernsehformat à la Maybrit Illner (Holly 2015), einem Weblog (Meiler 2018), erst recht in einer hochkomplexen Meso-Kommunikation wie bei Domke (2014) vorliegt. Es sind dies genau solche Strukturen komplexer Vernetzungen von Menschen und Dingen, wie sie neuerdings auch als Infrastrukturen beschrieben werden. Hier ist die Weiterführung des Kommunikationsformbegriffs bei Meiler (2018, 122 ff.) beispielhaft und sollte zur Kenntnis genommen werde. Das könnte manche zähe Missinterpretation endlich ausräumen.

    In Zeile 138 steht ein Verweis auf Hanitzsch 2007, in Zeile 910 wird auf Keppler 2015 verwiesen; beide fehlen im Literaturverzeichnis.

    Ich empfehle Überarbeitung.

     

    Zusätzliche Literatur:

    Berger, Peter L./Thomas Luckmann (1966): Die gesellschaftliche Konstruktion der Wirklichkeit. Frankfurt.

    Domke, Christine (2017): From the Space/Place-issue to Mediatization: On the Potential of the Concept of Communication Forms. In: Brock/Schildhauer, 53-73.

    Fiske, John (1987): Television Culture. London, New York.

    Holly, Werner (2004): Fernsehen. Tübingen.

    Holly, Werner/Peter Kühn/Ulrich Püschel (1986): Politische Fernsehdiskussionen. Tübingen.

    Jäger, Ludwig (2014): Medialität. In: Ekkehard Felder/Andreas Gardt (Hg.): Handbuch Sprache und Wissen. Berlin, Boston, 106-122.

    Meiler, Matthias (2018): Eristisches Handeln in wissenschaftlichen Weblogs. Heidelberg.

    Meyrowitz, Joshua (1985): No Sense of Place. New York, Oxford.

    Pflaeging, Jana (2017): Changing Potentials and Their Use: The Case of Popular Science Journalism. In: Brock/Schildhauer, 166-192.

    Reckwitz, Andreas (2007): Anthony Giddens. In: Dirk Kaesler (Hg.): Klassiker der Soziologie. 5. Aufl., München, 311-336.

    Schegloff, Emanuel (1988/89): From Interview to Confrontation: Observations of the Bush/Rather Encounter. In: Research on Language and Social Interaction 22, 215-240.

  2. Martin LuginbühlOctober 8, 2020 at 08:45Reply

    Lieber Herr Holly,
    haben Sie Dank für Ihr kritisches Gutachten. 
    Unsere Zurückweisung der sokratischen Was-ist-Frage in Bezug auf den Medienbegriff ist im Diskussionspapier noch nicht präzise genug formuliert. Sie kann in der Tat so gelesen werden, dass sie im Widerspruch zu unseren von Ihnen zitierten intensionalen Begriffsbestimmungen von ‚Medium‘ steht. Wir haben hier noch nicht hinreichend klargemacht, dass sich diese Zurückweisung auf die Extension des Medienbegriffs bezieht: Es führt u.E. in kategoriale Sackgassen, wenn man sich fragt, welche Objekte unter den Medienbegriff fallen. Dieser Problematik sind wir mit dem Zoom-Begriff, der für unsere Argumentation von zentraler Bedeutung ist, begegnet. Eine solche granulare Medienauffassung halten wir in Bezug auf empirische Analysen für sehr geeignet. Im Hinblick auf die Intension des Medienbegriffs ging es uns vor allem darum, die Idee von Medien als Verfahren der Zeichenprozessierung zu erläutern und an einem relevanten Beispiel, der Second Presidential Debate (2016), empirisch zu erproben. Uns ist aber durch Ihre Kritik klargeworden, dass die Formulierung, in der wir die Was-ist-Frage zurückweisen, missverständlich ist. Wir werden die entsprechende Textstelle überarbeiten. 
    Nach Ihrer Kritik werden wir Christian Stetters Medienauffassung in der überarbeiteten Version bei beiden Formulierungen (in Übersetzung) wörtlich zitieren und unsere Interpretation dieser Zitate deutlicher als solche kennzeichnen. Die Kippfigur, dass nämlich ein Verfahren, das in oder über einer Apparatur operiert, auf der anderen Seite auch als eine in Operation gesetzte Apparatur verstanden werden kann, ist eine Idee, die Christian Stetter vertreten hat und der wir uns anschließen. Wir möchten dies nun anhand einer Reihe von Zitaten aus „System und Performanz“ genauer belegen:
    „[D]ie Medialität des Mediums läßt sich nur fassen, wenn man dieses als in einem dinglichen Substrat operierendes Verfahren begreift, das aber selbst kein Ding ist, sondern eben Performanz.“ (S. 11)
    „So hat man in erster Näherung unter einem Medium das Sich-vollziehen einer Operation über oder in einem materiellen Substrat zu verstehen, über einem Apparat oder auch einem Konglomerat von Dingen, sodaß in diesem Vollzug etwas Wahrnehmbares von bestimmter Gestalt erzeugt wird: eine Sonate, ein Tanz, ein geschriebenes Wort, ein Stierkopf aus Fahrradlenker und -sattel.“ (S. 69)
    „Musik wird aus dem Eine-Geige-spielen nur, wenn und insoweit dieses gelingt. Das ist auf das Konto der Kompetenz des Spielenden zu rechnen, nicht auf das des Mediums. Dieses, das Operieren über einem Substrat oder auch: das in Operation versetzte Substrat ist für die Erzeugung des Mediatisierten […] zwar notwendige, nicht aber hinreichende Bedingung.“ (S. 70) –> Unter anderem hier findet sich die beschriebene Kippfigur: Ein Medium – so lesen wir diese Sätze – ist ein ‚Operieren über einem Substrat‘, und das heißt gleichzeitig: Ein Medium kann von der anderen Seite als ein ‚in Operation versetztes Substrat‘ begriffen werden.
    „Ein Medium – so können wir jetzt sagen – ist eine in Operation gesetzte Apparatur, sodaß durch diese Operation etwas, nämlich eine Darstellung von bestimmter Gestalt hervorgebracht wird. Medien sind – verkürzt gesprochen – symbolisierende Performanzen, genauer gesagt: das, was an der performance reiner Vollzug ist.“ (S. 74)
    „Auch ein solches Spur-setzen ist nur in einem Medium möglich, in einem über einer Apparatur welcher Art auch immer operierenden Verfahren.“ (S. 91)
    Gerade das erste und das letzte Zitat belegen, dass Stetter Medien als Verfahren in dem beschriebenen Sinne begriffen hat. Eine ins Englische übersetzte Zitation der beiden letzten Stetter-Formulierungen (S. 74 und 91) hintereinander, die wir in der Überarbeitung vornehmen werden, ist kein Zitationstrick, mit dem wir uns unberechtigterweise hinter einem großen Medientheoretiker verschanzen bzw. diesen als Autorität vorschicken wollen. Vielmehr wäre es u.E. wissenschaftlich nicht in Ordnung, diese auf Stetter zurückgehende Kippfigur nicht als seine Idee zu zitieren. 
    Auch Ihre kritischen Anmerkungen in Bezug auf den „New Institutionalism“ bzw. auf die Autorinnen und Autoren, die wir dort zitiert haben, beherzigen wir. In einer Überarbeitung werden wir diesbezüglich auf die lange wissenschaftliche Tradition (man könnte zu den Quellen, an die Sie uns erinnern, auch noch Gans oder Schudson ergänzen) wenigstens verweisen und so insbesondere das von Ihnen erwähnte Zitat von Shoemaker & Reese hinsichtlich seines Sachgehalts kritisch rahmen.
    Soviel für den Moment.
    Mit herzlichen Grüßen
    Martin Luginbühl und Jan Georg Schneider

  3. John BatemanOctober 9, 2020 at 10:43Reply

     
    When holism comes back to bite you:Comments on “Medial Shaping from the Outset”by Martin Luginbühl and Jan Georg Schneider
    John A. Bateman
    (Universität Bremen)
     
    (Note: this commentary is divided into two parts; the first is concerned with conceptual considerations, the second with implications for analysis and method. References are given in the second part.)
     
    Part I: Conceptual considerations
    Luginbühl and Schneider’s article pursues several important lines of development around the troubled, but nevertheless essential, concept of ‘medium’. Their detailed and thoughtful theoretical discussion succeeds well in beating a path from the early approaches of McLuhan and Kittler, through more refined work from Krämer, Jäger, Holly, Stetter and many others, up to the current time and the authors’ own position. The authors pay close attention to how the term ‘medium’ should best be situated with respect to notions such as ‘language’, ‘institution’, ‘material’, ‘technology’ and more, drawing on the perspectives of linguistics, media philosophy, and media linguistics. I find it particularly useful that this discussion now appears in English, since some of the most incisive theoretical and philosophical characterisations of ‘medium’ and ‘media’ produced since the 1990s, including work of the authors, have been conducted in the context of German media linguistics and, as a consequence, have not been as accessible to a non-German-speaking audience as they should be. All researchers and students of media communication need to engage with this development.
    The article is divided into two main segments, the first concerned with theoretical discussion and the second with illustrative analyses. The vast majority of the points made in the first half focus on the authors’ core claim that all communication is medially ‘shaped’ and that acceptance of this as a basic tenet offers a sounder basis both for theoretical development and for empirical analysis. In particular, as the authors state: “media contour the use of signs” (l. 315), while at the same time “individual and social use changes the media” (l. 316). Thus, although there is always a degree of freedom, “the media infrastructure shapes what we can do: it is always and inevitably part of meaning production” (ll. 279–281). The authors are quite meticulous in the presentation of their line of argument, and cycle back through the main point several times. This is because the claim that all communication is mediated, and so subject to considerations of medial effects, is apparently not universally accepted in media theory. Indeed, as the authors note, one traditional media theoretical distinction is drawn precisely between mediated, ‘technically’-enabled communication and ‘non-technical’, and thereby nonmediated, communication. Under this view, spontaneous face-to-face spoken interaction would be ‘media-less’ and so be bracketed from discussions.
    Luginbühl and Schneider argue, convincingly as far as I am concerned, that such a division is dysfunctional because it hinders productive comparative analysis of communicative practices that might fall, rather arbitrarily, on one side of a technical-nontechnical dividing line rather than the other or, worse, span both. The argument is thus very similar to that used in the introduction to our recent textbook on multimodality research (Bateman, Wildfeuer & Hiippala, 2017), where we point out that contemporary communication practices commonly ignore and transcend ‘traditional’ medial boundaries in just this way. Dividing the analytic cake respecting those boundaries will as a consequence often complicate practical analysis quite unnecessarily. This perspective is more or less forced on those who do multimodality research and there are important concerns here regarding the question as to whether it is, nowadays, possible to pursue ‘media’ research, in particular ‘media linguistics’, without placing verbal behaviour alongside accompanying semiotic systems. In any case, when faced with increasingly complex communicative situations, there appears little reason to accept earlier divisions and more productive characterisations of communicative situations must be sought. To illustrate their position, Luginbühl and Schneider select a very suitable area of analysis: televised presidential debates. Here face-to-face communication is clearly central since the presidential candidates must engage in largely spontaneous ‘natural’ interaction, but that interaction is nevertheless massively subject to highly technical mediation involving all of the characteristics and possibilities of live television.
    There are no doubt many working in media studies and media linguistics, particularly those orienting towards critical discourse analysis, who would be equally convinced that media shape communication and play an important role in the meanings that are created. However, Luginbühl and Schneider seek in addition to argue that this aim will be best served by adopting a more ‘holistic’ approach to the nature of ‘medium’. This holistic view combines several important facets and is proposed as a sound way of engaging in both theoretical refinement and empirical investigations of complex communicative situations. This then goes considerably further than rejecting any former restriction of medium to more technical or ‘hardware’ aspects. Much of the first part of the article is consequently concerned with removing conceptual boundaries that the authors consider unnecessary, so that existing terms or constructs, most centrally ‘medium’, can be progressively expanded in order to take up ever more of the explanatory load.
    It is here that more problematic aspects of Luginbühl and Schneider’s proposals surface. There is actually a considerable danger that analyses performed with respect to ‘medium’ construed as a relatively unstructured ‘whole’ unravel into looser collections of observations that are more difficult to turn into systematic empirical accounts. This becomes rather clear in Luginbühl and Scheider’s own example analyses of the presidential debate; the more complex the communicative phenomena addressed, the less guidance the supposed ‘holistic’ account provides. This is then an important concern for any media linguistics that seeks to engage with the diverse landscape of contemporary media. Luginbühl and Schneider suggest that analyses should focus on ‘medial procedures’ that address the “specific medium in question in its specific granularity” (l. 1085), but just what such a ‘specific medium’ might be becomes increasingly difficult to delimit.
    The authors’ progressive extension of the scope of ‘medium’ begins with the following connection between ‘media’, ‘language’ and ‘institutions’:

    “This view implies that there is no strict conceptual separation between mass media and semiotic media such as languages, because both are social institutions.” (ll. 160-162)

    To say that languages (‘semiotic media’) and TV (‘mass media’) are both institutions may be true, but to suggest that there is then ‘no strict conceptual separation’ appears a rather broad claim with questionable implications. The way in which these (ontologically) very different entities might be institutions is, I would maintain, completely different and so what this realisation demands is not to assume a lack of conceptual separation but instead to ask rather more precisely in what respects they might be considered institutions. What are institutions, or what do institutions do, such that both language and TV might be considered instances? Failing to draw a strict conceptual separation here is almost guaranteed to compromise analysis; I will illustrate this more concretely below with reference to the authors’ presidential debate examples.
    The reason that Luginbühl and Schneider take this path at all appears to be similar to that motivating their claim that all communication is mediated—that is: to draw comparisons across communicative situations or phenomena that might formerly have been separated. In the present case, this means that they can state that both TV and languages are subject to social constraints. But given that it is rather difficult (and artificial) to find any semiotic entities that are not subject to social constraints, this hardly adds information. It is certainly not a compelling reason to suggest there is no conceptual separation between these entities. Indeed, the argumentation here, and at several points in the paper, appears to be along the lines that if certain apparently distinct theoretical entities share some broader semiotic properties then those entities are not actually distinct (“no strict conceptual separation”) after all. This is a clear indication that the account is in need of further theoretical refinement.
    Lack of differentiation of this kind is then also shown in the authors’ discussion of ‘existence through use’. Luginbühl and Schneider argue, again I believe quite correctly, that use is critical to the existence of media (l. 408) and so a processual perspective is essential. This is one of the major hallmarks of the approach that the authors set out and the importance of seeing media as processual entities is emphasised again and again (e.g., ll. 411–412). But existence-through-use is surely an ontological condition for semiosis as such; semiotic entities need to be brought into existence by engagement by sign-users, they are not waiting in the world to be found, lying around like pebbles on a beach. And so, again, this cannot then stand as an identity criterion specifically for ‘media’: the property of emergence through use must apply to all semiotic entities (from languages, to political systems, to postage stamps) and, as a consequence, actually says rather little about the notion of ‘medium’. Now, Luginbühl and Schneider are quite right to emphasise the processual perspective and I am in this respect again in full agreement with them: as they write, there can be no ‘reification’ here. But this also echoes Peirce, Hjelmslev, Halliday, Giddens, Jäger and many others; it does not tell us anything particularly specific to the notion of ‘medium’.
    The authors’ account is then still left with the task of providing sufficient delimitation of what some ‘specific’ medium might be so that systematic and reliable analysis might be pursued. But Luginbühl and Schneider again seem to move in quite the opposite direction, apparently suggesting that, in fact, all semiotic entities, or at least ‘sign systems’, are media as well:

    “A specific feature of this definition [JB: of media] is the conceptual proximity of sign system and medium—the two terms describe one and the same multi-faceted phenomenon from different perspectives…” (ll. 243-246)

    Here it was unclear to me just what the single ‘multi-faceted phenomenon’ being described was. If it is ‘communication’, then this may be true, but is so broad that it offers little to take theoretical discussion further. Luginbühl and Schneider also appeal to an earlier characterisation offered by Schneider that suggests that that phenomena may simply be ‘sign use’. Sign systems then consider signs from the perspective of their ‘systemic and differential aspects’, in terms of schemas and rules, whereas media consider signs from the perspective of their materiality and processuality (Schneider, 2017, 45). Particularly the latter aspect raises substantial further questions, however: seeing media as medial procedures for processing signs does not of itself solve the challenge of finding out just what kinds of procedures this might involve. Again, for the development of this view as a contribution to analytic method, the notion of ‘medium’ still needs tighter characterisation. Instead, Luginbühl and Schneider invite further conflations by talking of describing sign systems “from the perspective of their mediality”, i.e., involving both signs’ materiality and their processuality, which then, logically, completes the circle to suggest that sign systems can be seen “as media” as well (ll. 247-249). Mediality then ends up as just the particular way in which some given medium processes material signs (ll. 253-255). What follows from this for analytic method is difficult to evaluate: it was actually unclear to me what precisely is intended to (and, indeed more importantly, what can) follow from the position.
    The problems raised by this grouping of phenomena become particularly evident whenever more detailed examples are considered and analytic precision is put to the test. I will show this briefly with respect to two final aspects of Luginbühl and Schneider’s theoretical discussion before turning to their illustrative analyses. First, there is their attempt to demonstrate that the notion of ‘communication form’, well known at least within German media linguistics, is simply redundant when one can work with a holistic view of medium. And second, Luginbühl and Schneider propose that medium is a good example of a ‘zoom concept’ (l. 266) capable of offering varying degrees of granularity for analysis as required. Both points reflect unresolved theoretical fault lines:
    1.     Communication forms have been proposed as a useful characterisation of ‘intermediate’ conventionalised categories such as emails, newspapers, radio programmes, and so on, which appear to hover midway between, on the one hand, particular technical media (computer messages, writing on paper, radio transmitters and receivers, etc.), and particular genres (e.g., factual informing of current events, entertainment, etc.) on the other. In Brock and Schildhauer’s characterisation of communication form, which the authors select for particular attention, ‘medium’ is consequently included as a proper ‘part of’ communication form (l. 334). This then indeed sees media more in terms of their material aspects combining the contributions of ‘trace-leaving’ technologies (for ‘technical media’) and biological capabilities (for ‘biological media’). In contrast, Luginbühl and Schneider argue that their ‘holistic’ view of medium already involves medial processes, sign systems, institutions, materials, etc., and so it would be more natural to structure things the other way round, with communication form becoming a (in their view redundant) part of ‘medium’ (ll. 388, 1082-1085). In many respects, however, this is a curious line of argument since it can only be made if one has already accepted that ‘medium’ does everything (and more) that communication forms do—in such a case, the concept might, indeed, be redundant. But this is precisely what Luginbühl and Schneider still need to demonstrate, rather than claim. Until analytic benefits can be shown, it remains questionable whether replacing a more finely articulated account addressing many important facets of communication with a ‘holistic’ concept lacking clear conceptual distinctions is a good strategy for further research and empirical study.
    2.     The tenability of Luginbühl and Schneider’s holistic view for effectively guiding analysis is not evidenced by their illustration of its ability to function as a ‘zoom concept’ either. Although presented positively as a flexible tool for supporting analysis, suggesting that analysis may then be adjusted to ‘fit’ any situation, it is far from clear whether the notion actually manages to structure situations to be analysed in a productive way. Even though variable focus is certainly a useful methodological mechanism for addressing a variety of more or less closely related communicative situations, when combined with a concept whose potential denotations have already been so broadly extended as Luginbühl and Schneider’s notion of ‘medium’, this very flexibility becomes a liability. The various entities or groupings of phenomena that Luginbühl and Schneider include under the umbrella term ‘medium’ are simply so heterogeneous that consistent relationships of ‘scale’ can barely be defined among them. Put simply: there are no clear dimensions along which might be ‘zoomed’ and this makes it difficult to apply the concept reliably as would be required for it to fulfil the methodological role that Luginbühl and Schneider suggest for it.
    It will be useful to unpack this latter point a little further. The authors illustrate zooming with the following example:

    “If the medium ‘spoken language’ is to be compared with the medium ‘written language’, the scopus is relatively wide and coarse; a comparison between face-to-face and telephone communication is narrower, and a comparison between landline telephony and mobile telephony is narrower still.” (ll. 268–273)

    At first glance, this might appear plausible: it seems that we are looking at broader and narrow communicative situations, which are then also (by conceptual conflation) different scales of media. In each case, there is no doubt that certain properties of the respective situations have an influence on the precise forms of communication employed: spoken language is known to differ from written language, speaking in a shared spatiotemporal context will differ from speaking ‘at a distance’ in relation to the use of deictics, gestures, etc., and using mobile phones will be more likely to lead to discussions of location, since this aspect of the situation will be variable. However, characterising all of these configurations as examples of more or less fine-grained media obscures the fact that the zooming relationship, such as it is, is working differently in each case because the situations described are actually semiotically very different.
    With spoken and written language, for example, we have a complete change in the material basis: the materialities of spoken language and of written language are disjoint. This has sweeping consequences for the semiotic systems that are, and can be, employed—indeed, the differences in affordances that follow from this are sufficiently substantial that models assuming that we have a single language system that is ‘simply’ expressed differently in the two cases have come to look increasingly naive (for critical discussions, see Krämer, Cancik-Kirschbaum & Totzke 2012 and van Leeuwen 2020 among others). It has become increasingly clear that both written language and spoken language involve highly complex constellations of cooperating semiotic resources and so each may well be characterised as ‘multimodal’ in its own right. More technically, we then need to say that written language relies on a set of cooperating semiotic modes operating with respect to one materiality and spoken language relies on a set of cooperating semiotic modes operating with respect to a quite different materiality. This predicts that there will be substantial differences between the way in which ‘language’, construed narrowly, will appear in the two cases because co-deployed semiotic modes tend to interact (cf. Winkler, 2008). Consequently, explaining those differences will require attention to just which semiotic modes ‘accompany’ core linguistic resources in each case. Spoken language and written language obviously share considerable portions of their semiotic potentials (e.g., much of their lexical, grammatical, semantic and pragmatic organisations), but their precise inter-relationship is, at this time, unknown and needs further empirical examination. One can say that there are partial congruencies across the semiotic modes of language as employed in written language and of language employed in spoken language, but whether these can both be characterised as further specialisations of a more general semiotic system shared between them or whether the relation is one of partial overlap needs considerably more research. Such research must, indeed, be high on the agenda of any ‘media linguistics’.
    The situation concerning the contrast between face-to-face spoken communication and speaking on a telephone is very different and so a different kind of granularity is at issue. Instead of the massive difference in materiality characterising spoken and written language, the material relationship here is relatively simple. Telephony can be characterised quite precisely as constructing a new materiality in which two dimensions of the materiality of face-to-face communication are removed: the first covers all but the aural components of that materiality, and the second is the perceptual material property of diminishing with distance. The contrasts that these material differences require of the deployed verbal semiotic mode are relatively small. All visually-supported forms (gesture, facial expression, etc.) are no longer available, but one can communicate at a distance. This is consequently a much simpler kind of contrast to the previous case, characterisable as a straightforward reduction in communicative bandwidth. This reduction must then be (and evidently has been) weighed against the broadening in social activities that are made possible when the situational constraint of co-presence for spoken interaction is no longer active. Both the contrast relationship between face-to-face communication and telephony and the relationship back to ‘spoken language’ as a presumed more general category are then quite different in nature to the contrast of the previous case. This means that any zooming in from language to either spoken language or written language is not the same as zooming in from spoken language to either face-to-face communication or telephony. Different semiotic processes are at work and these may be expected to have difference consequences both for analysis and for the kinds (and extents) of the variations in communicative expressions and practices that will be found.
    Finally, the change between landline and mobile telephony is different yet again. The same reduction in communicative bandwidth applies as in the previous case, but communication is no longer spatially ‘tethered’. This extends the social situations and activities that can be supported substantially, but leaves the materiality (more specifically, the canvas: Bateman et al., 2017, 213–217) unaltered. As a consequence, this does rather little to the spoken language semiotic system that operates in both cases. Differences in the use of the semiotic system will occur due to the different social activities that can be performed, but since language use is well established to exhibit variety according to situation (i.e., register) in any case, such differences are hardly surprising. Again, the motivations and sources of variation are of different kinds to that of the previous two cases, each involving different semiotic dimensions. In the first case, disjoint materialities mean that the relationships between deployed semiotic modes is one of broad (but not complete) congruence; in the second case, the relationships between deployed semiotic modes is a simple subset together with a particular collection of possible social activities; and in the third case, the difference is an extension of the social activities of the previous case.
    Grouping distinct semiotic configurations under a single, more ‘holistic’ label of ‘medium’ in fact makes it quite difficult to pick out precisely where observed differences of the above kinds could most appropriately be located. And this in turn makes it more difficult to predict and explain the kinds of differences that are observed. The notion of zooming then tells us little concerning whether we are dealing with variations in media or not. If, for example, we change the communicative situation of typical face-to-face spoken conversation slightly so that interactants are 100 metres apart, this impacts the materiality available for using the semiotic modes of face-to-face interaction considerably. Differences in language use will then be readily observable (for example, speakers might speak more loudly, enunciate more clearly, perhaps gesticulate more, and so on). Now such a situation is clearly a more specific situation than situations in general where spoken language might appear, so one has certainly ‘zoomed in’ on a more specific case: But has this changed the ‘medium’? I remain unsure from Luginbühl and Schneider’s discussion what the sufficient and necessary conditions on medium would be so that I could answer this question. If we extend things further, and constrain the interaction to take place between participants distributed on different sides of mountain valleys, then, with a few more stipulations, we might arrive at yodelling, which might well (by some) be considered a medium of communication. But we might also by such changes of granularity also arrive simply at shouting loudly and clearly, which (presumably) would not be a particularly promising candidate for a medium. The metaphor of ‘zooming’ therefore seems to suggest that any zoom settings might be taken, but this simply devalues the notion of medium still further since it is unclear why some arbitrary zoom setting should constitute a medium and why (presumably many) others do not.
    Among the clusters of concepts that Luginbühl and Schneider include within medium, some appear more useful for distinguishing media than others. This means that we can actually turn the discussion around and use zooming as a tool for picking out those distinctions that it would be useful to maintain. If distinct dimensions within the cluster concept ‘medium’ lead to non-congruent configurations when following paths of increasing or decreasing granularity, then those dimensions need to be formally separated, not conflated. For example, what appears to turn one class of interactions into a medium and another not is the sociocultural work that is put into its institutionalisation by a community. The more an entire network of institutionalised practices supports some activity in question, the more convincing a candidature for medium-status should become. With institutional practices we would then have at least some hold on criteria for delimiting media. Important then, however, is that if granularity or zooming is applied, it must be explicitly restricted to relative degrees of institutional granularity rather than being allowed to range over communicative situations in general. Only then might one expect shifts in granularity to have any predictive value or consistency.
    Another very different dimension within Luginbühl and Schneider’s notion of medium is materiality. Luginbühl and Schneider include material concerns alongside institutional aspects with relatively little discussion, reflecting the common practice of using the term ‘medium’ ambiguously both for characterising materiality (‘clay’, ‘oil and canvas’) and for characterising institutionalised communicative practices (‘mass media’). Although emphasising that medium cannot be reduced to materiality, Luginbühl and Schneider nevertheless maintain materiality as one of the concept’s hallmarks. But, since there is no guarantee, or logical requirement, that changes in granularity along the dimension of materiality would be identical (or even related) to changes in granularity along the dimension of institutionalisation, the two dimensions actually need to be maintained separately both theoretically and methodologically. In fact, this suggests that we now need to go further and explicitly rule out the notion of material or materiality from ‘medium’. Although a separation of ‘materiality’ and ‘medium’ along these lines may appear counter-intuitive given the long conflation, or at least close correspondence, of ‘medium’ and ‘material’, it is I think essential in order to prevent the resulting ‘superterm’ of ‘medium’ remaining too thinly stretched to really work. Institutional systems of communication, on the one hand, and materiality, on the other, simply pull in such different semiotic directions that combining them is deeply problematic. Coping with the resulting theoretical and conceptual difficulties has occupied much of the relevant literature for many years and this may well be a good indicator of the need for substantial re-theorisation. Discussions typically either deal with medium as materiality, with institutional constraints somehow riding on its coattails and analytically marginalised in practices or background knowledges of various kinds, or deal with institutions, with a corresponding lack of analytic precision concerning the concrete details of unfolding communicative artefacts or performances. Medium should then, arguably, focus more specifically on institutionalised communicative situations and the ways in which those institutions constitute and maintain themselves in communicative practices. The semiotic concerns of materiality are quite different and, as I will now suggest, can be put to good use to tighten empirical methodology.
    (cont’d in part II)
     

  4. John BatemanOctober 9, 2020 at 10:48Reply

     
    When holism comes back to bite you:Comments on “Medial Shaping from the Outset”by Martin Luginbühl and Jan Georg Schneider
    John A. Bateman
    (Universität Bremen)
     
    Part II: Implications for analysis and method
    My discussion so far has been primarily theoretical, arguing that working towards a ‘holistic’ understanding of media in the manner that Luginbühl and Schneider propose overburdens the term to the extent that its internal coherence is threatened and its ability to serve as an incisive methodological tool for empirical analysis compromised. In this final part of the commentary, I show this more concretely with reference to the analyses that Luginbühl and Schneider offer of the second televised presidential debate between Donald Trump and Hillary Clinton in 2016. These analyses cover three main areas: discussion of the social interaction, including use of forms of address and gaze behaviour, discussion of the setting as a mock ‘town hall’ debate to simulate a real question-and-answer context, and discussion of the editing techniques employed for the televised transmission. Although Luginbühl and Schneider’s main goal is to show medial effects on everyday conversational interaction, most of the details of the verbal interactions shown would also occur in a much broader range of contexts than televised presidential debates. Analysis of such situational variation is, of course, interesting and revealing but Luginbühl and Schneider’s linking this so closely with the discussion of medium and its definition is less convincing. Language use is well documented to be sensitive to, and to co-construct, social contexts (registerial variation and situational dependence) and in this respect political debate is no different, televised or not; the fine-grained linguistic analysis of interaction in such contexts is of course an established research field of its own.
    Luginbühl and Schneider’s need to connect this kind of variation directly to medium then appeared to me to distort the account in several places. At one point the authors talk of the “medium’s needs for entertainment and balance” (l. 882): it is unlikely that the medium (televised debate?) has such a need, although the institutions of journalism and entertainment certainly do. This conflation is also evident in the authors’ characterisation of the situation:

    “The candidates’ gaze work, the way they use the studio room, how they walk, align and disalign their bodies, are all shaped by the medium of television. As in language, the effect of the medium is not secondary. The medium itself influences the behavior of the persons on screen from the outset and creates its own reality of the conversation. A reality that cannot be experienced at the location itself.” (ll. 1047–1053)

    Here one can safely say that the effect of the situational context is not secondary and that that context influences all aspects of the behaviour of the participants, as it always will. But it is by no means clear to me that this can be relatable to the ‘medium of television’ so directly—at least not without already conflating medium, institution, and social fields of activities as already critiqued above. The authors have, up until this point, still said rather little then that would help determine just wherein the ‘specificity’ of a ‘medium’ would lie.
    Luginbühl and Schneider’s reference to the medium ‘creating its own reality’ is very important for any critical media analysis, but is then not really addressed by the parts of their analysis focusing on verbal interaction. Those analyses reflect how power-negotiating interactional verbal discourse (including gesture, gaze, turn-taking, and so on) unfolds in general. When such interactions are viewed (whether on TV or at the back of the studio among the audience, or in a real town-hall), audiences will use their knowledge and experience of social face-to-face interaction in the genre of political debate to understand what is occurring and to draw conclusions concerning the debaters. This ‘reality’ can certainly be experienced at the location itself and so is, arguably, perhaps not even central to Luginbühl and Schneider’s point in any case. These characterisations of unruly attempts to gain the floor, addressing both local and remote audiences, unusual topic developments, etc. are all typical of the verbal interaction of political discourse and even sometimes organisational meetings of other kinds where power is being negotiated, not of the ‘medium of television’.
    This means that Luginbühl and Schneider’s methodological call to address the specifics of the medium in question only really gains weight in the third discussion area, where some of the actual ‘specifics’ of a televised broadcast come into play. Only here is there discussion of a medially created ‘reality’ that can, indeed, only be experienced through use of the medium (i.e., by watching TV). To show this, Luginbühl and Schneider extend their analyses into areas that are clearly highly multimodal—even though the interactional behaviour of the participants analysed in the earlier examples was already highly multimodal (gesture, gaze, body posture, proxemics, etc.), this situation is extended significantly when considering camera editing techniques, staging, and so on. Analysing such multimodally rich complexes raises considerable challenges and it is here, even more than was the case for the verbal analyses, where strong methodological guidance for empirical study is essential. Such guidance is not forthcoming from Luginbühl and Schneider’s ‘holistic’ view of medium, however—indeed, important guidance appears even to have been compromised by virtue of the weakened conceptual distinctions available.
    This is then a crucial point for any ‘media’ linguistics that seeks to go beyond studies of verbal interaction as Luginbühl and Schneider argue is necessary. Indeed, the fact that the analytic and methodological problems raised so far become so apparent when forced to deal with the many non-verbal aspects of television is particularly telling. Whereas an engagement with the multimodality of a communicative situation can perhaps, at least to a certain extent, be marginalised when restricting attention to verbal phenomena, this is absolutely impossible when moving to the affordances of other media. The ability of genuinely non-verbal audiovisual presentation strategies to co-construct their own ‘realities’ is well known in film studies and critical discourse analyses of audiovisual media alike. Nevertheless, there is still a considerable need to develop sound empirical methods for dealing with these communicative situations more effectively. An appropriate characterisation of medium and its relationship with institutions and materialities would certainly help with this, but the possibilities for this within Luginbühl and Schneider’s account are limited. It is also likely that a more well-founded position on multimodality in general will be of equal benefit for verbal interaction studies also.
    Televised presidential debates in particular have long played an important role in U.S. politics and so it should be no surprise that their ‘medial construction’ has also been the subject of study for a considerable time. Rather more surprising is that Luginbühl and Schneider cite almost nothing from this body of work, perhaps because its orientation is not usually to the fine details of linguistic or conversational interaction, but rather on the recipient effects of the diverse audiovisual techniques that the medium makes available. As noted by Stewart, Eubanks, Dye, Eidelman & Wicks (2017), for example:

    “The subtle, yet highly important production choices that we usually associate with TV and movies—duration and type of camera shots, angles, and camera movement—play a major role in our perceptions of candidates (Bucy & Newhagen, 1999; Grabe & Bucy, 2009). This is especially the case in the presidential debates where the predominant camera shot broadcast is the side-by-side split screen shot, with both faces focused on.” (Stewart et al., 2017, 548)

    Here we can see that media influence, and even the co-construction of meanings through medial practices, is not only accepted as Luginbühl and Schneider argue, but also actively worked with as the primary object of empirical study. Luginbühl and Schneider turn their attention in their third area of illustrative analysis to precisely these aspects of camera editing and positioning, uses of split-screens, stage setting, movement of the candidates, and more. The omission of references to this work in Luginbühl and Schneider’s discussion is then doubly unfortunate. First, precisely the phenomenon that Luginbühl and Schneider pick out for more focused illustrative analysis, ‘split-screens’, has already been extensively investigated and its effects on audiences’ evaluations of candidates empirically evaluated; this could have been used to bolster claims more effectively and set research agendas: for example by inter-relating such results with fine-grained verbal interaction analysis. Second, a more explicit engagement with this tradition would also have brought home with considerable force the levels of empirical detail that are now required if critical analysis from media linguistics, or from any other field, is to make useful contributions for understanding this kind of media.
    As cited above, Luginbühl and Schneider suggest, somewhat uncontroversially given the literature, that technical ‘filmic’ manipulations have significant effects that create a reality “that cannot be experienced at the location itself” (l. 1053). This needs to be seen, however, as an essential property of the medium and be analysed as such—Luginbühl and Schneider seem to consider the medial presentation as a ‘distortion’ of the verbal interaction, rather than as an entity in its own right as their own methodology would recommend. The authors even talk of the creation of “impossible” viewpoints (via the use of split-screens), as if the ‘unmediated’ natural interaction is the defining metric. Audiences are long used to filmic presentations that do not rely on indexical spatial relationships. From the perspective of the analysis, therefore, we see a rather logocentric viewpoint that offers detailed accounts of the verbal interaction but only more sporadic observations concerning audiovisual presentation strategies. As a consequence, although the points that Luginbühl and Schneider make concerning these aspects of the design of the televised event are all interesting and invite discussion of how the televised event is responsive to the various institutional logics at issue, the analysis itself remains more ‘opportunistic’ with respect to which aspects of design are focused upon, precisely because of the lack of methodological guidance that the loose notion of ‘medium’ provides. What is analysed appears then too often to remain what strikes the analyst as being of interest.
    What must be placed more at the centre of attention, therefore, is precisely what any supposed ‘medial shaping’ is and how it might work. And, for this, it is crucial that larger samples of data are systematically analysed, which itself demands that the situations analysed can be reliably segmented and the diverse sources of influence on production and interpretation tracked. It is also important here that relevant trajectories of development and change across ‘media’ be recognised, which requires in turn that a deeper understanding of the workings of such trajectories be developed. It is precisely these multimodal phenomena that reflect most closely the specific potentials of this particular media manifestation of the institutional logics of entertainment, journalism and politics that Luginbühl and Schneider are striving for. And, to achieve this, it is necessary to articulate more clearly the different sources and constraints conditioning the observable variations.
    Split-screens offer a good case in point. The use of split-screens has a long history in (narrative) film and much is known both about the meanings they are typically employed for and how audiences interpret them. How, however, do we formally relate such usages to the medially very different situation of the live televised presidential debate? On the one hand, there is a carry-over of production practices from film school and viewer experience to television production; describing this carry-over needs to engage with practitioners. On the other hand, we have a continuity in the materiality that is available for making meanings, from film through to television. Just as was explained above to motivate the (semiotic) differences between spoken language and written language, attention to materiality is key. The materiality of early television production did not support showing two images at the same time and so this presentational strategy from film was not available; now it is possible and so the materiality of television has changed accordingly. The simple availability of a material resource does not, however, determine how that resource may be used, although there are natural constraints. The functions of split-screen effects in film generally involve comparison, contrast and, more optionally, simultaneity. These are, following the account of multimodality given in detail in Bateman et al. (2017), potential discourse interpretations for the materially realised audiovisual form. The use of split-screens on news-oriented television, while naturally drawing on this history of interpretative practices, has in turn developed a more restricted range of (institutionally motivated) potential functions whereby what would formerly need to have been shown in sequentially distinct ‘reaction shots’ can now be made continually accessible to the audience. An important difference to reaction shots is that sequentiality is not enforced: that is, whereas it is barely possible not to read a close-up of a facial expression following a statement made by someone as a reaction to that statement, in the split-screen version, it is left to the activity of the recipient to read some visible bodily action as a reaction to what is occurring in the other half of the screen. Institutionally, therefore, the split-screen might be seen as less manipulative as there is no material possibility of ‘visually asserting’ (by sequence) that something is a reaction to something else.
    It is, however, precisely this continuous visual access that leads to problems, of course. Luginbühl and Schneider pick out particular cases of this in their analysis but the more general treatment of the issue remains thin. Whereas professional anchor-people interacting for a short period of time can look attentive and offer appropriate nods and smiles, this is difficult to maintain over the course of a debate. Any action, however unplanned and insignificant, may become subject to recipient interpretation—particularly when shown in close-up. This is due both to the natural tendency of behavioural expectation violations to attract attention (Gong & Bucy, 2016) and to the learned media consumption habits of narrative film, where if something is shown it is automatically considered to be potentially significant (something akin to a visual version of relevance theory). In split-screens, therefore, where close-ups may be shown continuously of reactions, the likelihood of mishaps is very high and there are many reports both in the scientific literature and in the press of such cases occurring.
    This can now all be brought together using the distinctions I argued above to be necessary semiotically for characterising the notion of ‘medium’ productively. I suggest that this allows for far finer-grained and more tightly guided investigations and predictions of the effects of any communicative situation studied. First, the medium, as set out above, may be considered to be the institutionalised practices for information distribution and creation, such as, in the present case, live television. It is the medium that interfaces directly with broader fields of social activity, such as journalism, entertainment, etc. Moreover, it is these broader activities that set priorities for how the possibilities offered by some medium are to be deployed and developed. This is by and large still compatible with Luginbühl and Schneider’s framework. Going further, the meaning-making aspects of media are then carried by semiotic modes (written language, spoken language, diagrams, the moving audiovisual image, etc.), each of which offers resources for manipulating particular materialities and assigning interpretations to the material distinctions formed. This ‘moves’ issues of materiality away from the notion of ‘medium’ and places them ‘within’ the account of semiotic modes. One motivation (and consequence) of this is that very similar, even identical, materialities may be deployed by quite different media, such as in television and film. Finally, and absolutely crucially for pursuing detailed investigations, explicit characterisations of the materiality being deployed grounds empirical analysis—whereas it may not be entirely clear just which social institutions may be at work, the materiality that is present is always maximally concrete and therefore offers a much needed more robust starting point for analysis.
    The situations that Luginbühl and Schneider describe can now receive more incisive descriptions. When they write, for example:

    “Final control over the meanings that are broadcast therefore lies with the medium—that is, with the media institution’s picture director…” (ll. 902–905)

    we can see that it is actually quite questionable whether (and how!) ‘meanings’ are broadcast. Since semiosis requires materiality, it is instead an audiovisual text that is broadcast in which a variety of perceptible material cues are manipulated in order to invite certain discourse interpretations on the part of viewers. Moreover, for appropriately critical analysis, it is important to note both that such cues can be picked up in radically different ways depending on the media experience, cultural orientation and knowledge of a recipient (i.e., different discourse interpretations will be reached), and that interpretations may actually be accessible to neither the ‘picture director’ nor the recipients themselves. Scheufele, Kim & Brossard (2007), for example, report on systematic polarisation effects where the bare use of split-screens regularly reinforced pre-existing negative opinions held by audience members concerning a debate participant. One simple explanation for this latter result would be that negative interpretations of reactions by someone about whom one already has a negative impression offer a relatively low-cost way of achieving overall discourse coherence—that is, if one expects negative behaviour then interpreting cues negatively is easier than interpreting them positively. Stewart et al. (2017) also report significant effects on candidate trait judgements depending on whether split-screen or solo camera shots were used for the second presidential debate between Trump and Clinton in 2016. The effects were stronger for Trump, although in both cases the solo shots led to more differences in trait judgements than were the case with the split-screen. For this reason as well, detailed study of such communicative situations is very important so that unintentional manipulations of the impressions that audiences take away from such media events can also be recognised and, ideally, avoided (cf. Stewart, Eubanks & Miller, 2019). Final control therefore never lies with the medium or with the media institution, although skilled producers (and, even more, film directors) may certainly strive to approximate this. The starting point for analysis must nevertheless always be the materiality  and the distinctions that are employed in that materiality according to the semiotic modes in use.
    There are also important differences concerning just which agents get to manipulate the materiality being deployed. Luginbühl and Schneider discuss cases where Trump moved to visually prominent positions where he could be picked up by a camera that was also showing Clinton’s response to a question. Luginbühl and Schneider point out that although this latter shot was claimed to be a strategic move by Trump, it was also the decision of the production team to select that camera view at that time so they might also be considered at least co-responsible (ll. 1035–1039). Nevertheless, to address this issue properly it would also be necessary to know both the camera selection strategies and Trump’s knowledge of them. Ensuring that sight-lines are maintained and appropriate positioning with respect to cameras constitute standard resources of the semiotic modes of staged film-making. If ‘actors’ know that they will be picked up in certain positions, then moving to those positions is also a way of co-determining the final media product. Thus, although the simple view would be that the decision is made by the picture director, if filmed subjects know the camera-selection strategies, then the entire situation is reversed: the picture director and camera selection technology becomes the manipulable material of the actor. Similarly with close-up views of emotional responses: if the filmed subject is aware that the response will be picked up it can be produced appropriately. It will not be possible for viewers of the final media product (nor analysts of the final media product) to know which of these situations actually held without further information.
    It is also by no means possible to simply attribute physical staging properties to the ‘medium’. The fact that the chairs of the debaters were placed in a particular way belongs strictly to the semiotic mode of staging, which has variants in film and theatre as well as television. The material distinctions of this semiotic mode are constituted by designated ‘places’. What actors (in a broader sense) then do with these resources is still relatively free. They will manipulate the materiality as far as they can to achieve their own goals, which may, of course, also be the goals that they adopt within the scope of institutional fields of activity. The physical situation will effect and support certain patterns of movement rather than others as it always does but this by no means demands an institutionalisation that would warrant characterising it as part of the medium; in the next debate, the chairs could be positioned differently without ‘changing’ either the medium or the institutional goals that the medium is being used to pursue. The medium sets neither the staging itself nor the actors’ behaviour. The staging and behaviour may be interpreted with respect to such institutional fields, as Luginbühl and Schneider suggest, but that interpretation necessarily takes place at several removes.
    Precisely the same applies to several of the further constraints discussed, such as limiting the speaking time for responses to two minutes (l. 707). This is again not directly constitutive of the ‘medium’ but is instead a relatively arbitrary production constraint imposed on the communicative situation, similar to the adoption of rules in a game. Unless this were to become an ‘institutionalised’ practice in its own right, there is little reason to consider it ‘part of’ the medium (although it might well already be considered a part of a subgenre of debate available in many media: again, an empirical issue). The manner in which it will effect linguistic performance is relatively predictable as well (free production under time pressure) and this again is only indirectly responsive to the institutional fields of social activity being performed. In both cases, then, Luginbühl and Schneider appear to be trying to make ‘medium’ do rather too much work, which is, in fact, a rather common tendency in accounts that centre on ‘medium’ without the support of complementary notions such as ‘genre’ and ‘semiotic modes’ (cf. Bateman, 2017). In contrast, both the two minute response time and the positions of chairs can quite naturally be considered further specifications and restrictions of the materiality available, resulting in clearer predictions and expectations concerning the (range of) behaviours that will ensue.
    Conceptually separating medium, semiotic mode and materiality in this way is then a further means of avoiding potentially too deterministic views of the relation between media and actors (ll. 294-298). As Luginbühl and Schneider emphasise:

    “Even mediality as a whole is not a sufficient, but only a necessary condition of media usage. The crucial point is this: for scientific analysis, a distinction must be made between (a) mediality, i.e. the possibilities of the medium, (b) the sign system(s)/modes employed, (c) the communicative practices (language games in a Wittgensteinian sense) and (d) the skills of the players.” (ll. 298-304)

    Both (c) and (d) appear unproblematic. This commentary has consequently focused particularly on the distinction between mediality and sign systems and how this has been constructed by Luginbühl and Schneider. The expansion of ‘medium’ has been suggested to undermine the distinction rather than make it more usable in scientific analysis. Consider then, as a last example from a completely different domain, the fact that online chatting does not support interruptions: where is this to be situated and accounted for in a description? For accounts that only have ‘medium’ available as a functioning theoretical construct, there is little choice. Within the framework sketched here, there is no need to assign this property to medium as it is already inherent in the low temporal granularity of the materiality provided. It is, after all, unlikely that this property was ‘designed in’ on institutional grounds and could well disappear with further revisions of the supporting platform without necessarily changing the medium. In a related field, producers of semiotic technologies such as Microsoft’s PowerPoint, Adobe Illustrator, etc. are best seen as designing first and foremost materialities, rather than ‘media’. It is, after all, questionable whether a medium, as a broad institutionalised sociocultural system of communicative practice, could actually be ‘designed’ at all. It is the social fields such as politics, economics, journalism, art, etc. that set priorities for what such institutionalised media will seek to ‘communicate’ and this may well push (and be pushed by) the development of supporting materialities in various ways—but the relationships here only operate indirectly and nondeterministically via the situated deployment of semiotic modes and it is these that offer specific frameworks for data annotation and interpretation.
    In short, the characterisation provided by Luginbühl and Schneider shows convincingly that a beneficial understanding of ‘medium’ should engage with institutional concerns and characterise such concerns in terms of processes of sign production and interpretation. Its way of relating medial processes with actual phenomena still appears, however, to be too focused on language and so does not give sufficient weight to the key role of materiality and the shaping of materiality by semiotic modes—which will then also influence co-present verbal phenomena. When moving in on the phenomena of communication from a perspective not tied within linguistics, additional motivating forces for potential frameworks become visible. These argue that a rather different decomposition and interrelationship of theoretical components are necessary. As a consequence, from the broad perspective of multimodality, particularly as proposed in Bateman et al. (2017) that I have drawn on repeatedly here, Luginbühl and Schneider’s discussion still maintains an inherent vagueness concerning ‘mediality’ and ‘materiality’. This ambiguity is present because of a relatively under-developed view of signs and materiality and their necessary connection within semiotic modes. Thus, rather than saying that even face-to-face spoken communication ‘has its own specific mediality’, the view from multimodality assumed here would say instead that face-to-face spoken communication has its own specific materiality, not mediality. After all, the semiotic mode(s) of face-to-face spoken communication can be deployed in many medial processes and so does not have its ‘own’ specific mediality at all. Instead, in each case, it is the deployment of the mode that may shape and be shaped by institutional concerns. In addition, specifically for media linguistics, one needs to consider the effects of semiotic modes that are co-deployed alongside linguistic modes in specific medial contexts as well.
    Precisely the same (semiotic) description of the necessary relationships between theoretical components applies to all semiotic modes and it is this that establishes a robust foundation for research that includes, but which also extends beyond, verbal language. Whether some semiotic mode is bound into some medial processes is then strictly an empirical issue to be assessed with respect to actual sociohistorical institutional processes. Performing the empirical work necessary for investigating this benefits considerably from the additional conceptual distinctions drawn between media, materiality, and semiotic modes since, only then, does it become possible to characterise the semiotic contributions of semiotic modes with sufficient integrity to show both their meaningful variation in the service of diverse medial configurations as well as their paths of reuse and transformation across such configurations. Conceptually conflating media, signs and materiality makes such tasks of analysis far more difficult than they already are and so, methodologically, is to be dispreferred.
    References
    Bateman, John A. 2017. Triangulating transmediality: a multimodal semiotic framework relating media, modes and genres. Discourse, Context & Media 20(Supplement C). 160–174. doi:10.1016/j.dcm.2017.06.009.
    Bateman, John A., Janina Wildfeuer & Tuomo Hiippala. 2017. Multimodality – Foundations, Research and Analysis. A Problem-Oriented Introduction. Berlin: de Gruyter Mouton.
    Bucy, Erik P. & John E. Newhagen. 1999. The micro- and macrodrama of politics on television: Effects of media format on candidate evaluations. Journal of Broadcasting & Electronic Media 43(2). 193–210. doi:10.1080/08838159909364484.
    Gong, Zijian Harrison & Erik P. Bucy. 2016. When style obscures substance: Visual attention to display appropriateness in the 2012 presidential debates. Communication Monographs 83(3). 349–372. doi:10.1080/03637751.2015.1119868.
    Grabe, Mary E. & Eric P. Bucy. 2009. Image bite politics: News and the visual framing of elections. New York, NY: Oxford University Press.
    Krämer, Sybille, Eva Cancik-Kirschbaum & Rainer Totzke (eds.). 2012. Schriftbildlichkeit. Wahrnehmbarkeit, Materialität und Operativität von Notationen. Berlin: Akademie Verlag.
    Scheufele, Dietram A., Eunkyung Kim & Dominique Brossard. 2007. My Friend’s Enemy: How Split-Screen Debate Coverage Influences Evaluation of Presidential Debates. Communication Research 34(1). 3–24. doi:10.1177/0093650206296079.
    Schneider, Jan Georg. 2017. Medien als Verfahren der Zeichenprozessierung:Grundsätzliche Überlegungen zum Medienbegriff und ihre Relevanz für die Gesprächsforschung. Gesprächsforschung – Online-Zeitschrift zur verbalen Interaktion 18. 34–55. http://www.gespraechsforschung-online.de/fileadmin/dateien/heft2017/ga-schneider.pdf.
    Stewart, Patrick A., Austin D. Eubanks, Reagan G. Dye, Scott Eidelman & Robert H. Wicks. 2017. Visual Presentation Style 2: Influences on Perceptions of Donald Trump and Hillary Clinton Based on Visual Presentation Style During the Third 2016 Presidential Debate. American Behavioral Scientist 61(5). 545–557. doi:10.1177/0002764217707621.
    Stewart, Patrick A., Austin D. Eubanks & Jason Miller. 2019. Visual priming and framing of the 2016 GOP and Democratic Party presidential primary debates. Politics and the Life Sciences 38(1). 14–31. doi:10.1017/pls.2018.16.
    van Leeuwen, Theo. 2020. The new visuality of writing. In Hartmut Stöckl, Helen Caple & Jana Pflaeging (eds.), Shifts towards Image-centricity in Contemporary Multimodal Practices Routledge Studies in Multimodality, 64–85. London and New York: Routledge.
    Winkler, Hartmut. 2008. Zeichenmaschinen: oder warum die semiotische Dimension für eine Definition der Medien unerlässlich ist. In Stefan Münker & Alexander Roesler (eds.), Was ist ein Medium?, 211–222. Frankfurt am Main: Suhrkamp Verlag.
     

Leave a Comment