Sensory communication in YouTube reviews: The interactional construction of products

Introduction

Online consumer video reviews are a comparatively new phenomena that have emerged only in the last 10 years but that are already fundamentally changing everyday consumption practices (Alexander et al., 2018; Graham, 2018: 1). Scholars have been slow to turn analytic attention to video reviews (Blank, 2006), with researchers only recently taking an interest in the structures of these influential discursive forms (Bhatia, 2018; Jaakkola, 2018; Jorge et al., 2018). My particular interest in this article is in product reviews on YouTube, which, along with Amazon (Skalicky, 2013; Vasquez, 2016), has become one of the most important consumer platforms for finding out about products. More specifically, I am concerned with the ways that users produce objectified ‘symbolisations’ (Streeck, 1996) of products through descriptions that articulate/construct sensorial experience. YouTube reviews are a particular kind of ‘fractured ecology’ (Luff et al., 2009) where the presenter makes a video for a generalised other that can be responded to textually through comments and ‘likes’. There is no gestural ‘interaction’ between the participants, just a one-way sending of a multimodal message. Interactionism has a long-standing interest in how people manage technologically mediated actions (Heath et al., 2000, 2003) and I draw on these frameworks to explore the communicative practices of review work.

Existing research in this area has heavily emphasised that YouTube is not a neutral space of democratic citizenship, but a complex economic systems tied to capitalist consumerist ‘discourse practices’ (Jorge et al., 2018). The reach of consumption discourses can be found in the seemingly ubiquitous nature of ‘review work’ on YouTube, which can be found across a wide range of overlapping YouTube genres including parodies and re-hashes of existing videos (Burgess and Green, 2013; Leppänen and Häkkinen, 2012), room tours (Rebane, 2019), unboxing videos (Nicoll and Nansen, 2018), reaction videos (Kim, 2015), decluttering videos (Zappavigna, 2019) and tutorials (Bhatia, 2018). Reviewing may not be the central aims of all videos, but it is easy to find examples of vloggers commenting on the products they are engaging with. The majority of YouTube reviews – whether formally positioned as such or not – are not ‘institutional’ in that they are not produced by private companies, media outlets or other organisations (Jaakkola, 2018). However, review work is often connected to online or ‘bricks and mortar’ sales outlets, carefully staged and guided by production companies (Hou, 2018; Jorge et al., 2018; Nicoll and Nansen, 2018), with reviewers frequently paid or incentivised for their work (Gozegir and Gocer, 2018; Kong et al., 2018).

Following a similar emphasis in research on face-to-face and television shopping (Bucholtz, 2000; Cook, 2013; Niemi and Hirvonen, 2019; Soar, 2009), one of the dominant interests in YouTube reviews is with how reviewers manage sales as a discursive act. Jaakkola’s (2018) study illustrates how reviewers use humour and irony in ways that are, she suggests, reminiscent of theatrical ‘revues’ rather than conventional product reviews. Bhatia’s (2018) research on YouTube makeup artists also shows the complexity of discourse practices found in reviewing, illustrating how one vlogger shifted between traditional ‘teaching’ discourses – where viewers are shown how to achieve particular makeup techniques – to strategies that were directed to selling products. Jorge et al.’s (2018) study of one beauty vlogger also shows the difficulties of negotiating ‘discursive strategies that mesh intimacy and commercialism’ (p. 92). Such performative strategies have been characterised as part of ‘semiotic-capitalism’ and ‘digital-capitalism’ (Zappavigna, 2019), where discourses of consumption and materiality inform the ‘subject positions’ that people adopt.

These preliminary studies are valuable as they illustrate some of the tensions within YouTube review work. To date though, no research that I am aware of has examined closely the communicative practices of product description and the enactment of objectified accounts of products. My interest in this article is in drawing on sociological perspectives on the senses to explore how reviewers transform objects from ‘disembodied things’ into sensorially rich and interactionally embedded products.

Methods

The initial sampling frame for this study involved identifying particular categories of sense work – smell, taste, hearing, vision and touch – and searching for examples of reviews that dealt with those categories. One of the key conceptual points in the sociology of the senses is that these conventional categories of sensing are cultural constructs that potentially over-simplify our lived experiences of the world, which may involve experiences that do not fit well into these rough categories (Howes, 2006; Paterson, 2007; Pink, 2010; Vannini et al., 2011a; Gibson and vom Lehn, 2019). However, perhaps because such categories are so culturally enshrined, most studies of the senses do use them as the basis of their work (Gibson and vom Lehn, 2019). In the case of this study, I am interested in how conventional categories are used, and I too use them to inform the sampling and the organisation of the analysis to come.

I decided to look at reviews in English and Spanish in order to include non-English data in the analysis, which is a common limitation of interactionist work. I chose to focus on channels of more experienced vloggers, defined here as users/channels that had uploaded more than 80 videos and that had been posting for at least 3 years. I focussed on reviews where the senses would be of relevance, choosing wine tasting, perfume reviews, musical instrument reviews and mobile phone reviews (see Table 1). Wine tasting is a complex and highly ritualised practice that involves visually assessing, tasting and smelling. Perfumery is closely focussed on aroma and on the visual appearance of perfumes and occasionally on the tactility of packaging. Musical instrument reviews tend to focus particularly on the sound and the feel of instruments but also on aesthetic qualities such as how they ‘look’. Finally, mobile phone reviews focus on the ‘feel’ of the products, as well as sound quality, usability and visual aesthetics. In each of these areas of review practice, multiple senses are at play (see Table 1) and my analysis looks at the ways that these are articulated in the process of reviewing the products. There are differences in the extent to which sensorial experience are the focus of attention in the reviews. In wine tasting and perfume reviews, for instance, the articulation of sense experience is central to the entire review, while in mobile phone reviews, senses experience only occasionally emerges as a topic, and usually in relation to very specific aspects such as the ‘feel’ of the product or the sound of the speakers. I decided to include mobile phones as a less obviously ‘sensory’ category of review work in order to avoid treating sense work as only being present in special types of reviews.

OPEN IN VIEWER

Table 1.

Sampling framework for the videos analysed.

Type of product being reviewed	Senses most frequently present in reviews	Number of videos identified	Number of videos transcribed as a timeline	Videos transcribed as a multimodal transcript	Videos presented in article
Wine tasting	Smell, taste, vision	18	8	2	Vinistas – Milenramas Joven Review
Perfume	Smell, taste, vision	24	8	2	Brookln Fragrance Lovers – Terre d’Hermes Parfum Review
Mobile phones	Sound, touch, vision	27	8	2	Ryan Moore – LGG7 ThinkQ Review
Musical instruments	Sound, touch, vision	26	8	2	ThePianoForever – Steinway O Review

The first phase of analysis involved producing a timeline transcription to outline the general structure of each review. After this, I selected a subset of eight videos for detailed transcription through established techniques in EMCA (Heath et al., 2010; Jefferson, 1984) and wrote a narrative account of each. The narratives provided detailed analysis of the communicative work involved in the reviews, which I subsequently cross-checked with the wider data set, finding that they provided a sufficient elaboration of the interactional work across the data. The transcripts include screenshots of physical actions along with textual interpretations of the talk (see Appendix 1 for a description of the transcription symbols used in the data). Due to the restrictions of space and to avoid repeated descriptions of different contexts, the discussion here focusses on just one video transcript from each of the review types. The analytic points being made, however, relate to phenomena found across the data set.

Through the analysis, a basic distinction between types of sense work became evident and this informs the organisation of the data discussion to follow. I distinguish between ‘unavailable sense’, which the viewer has no access to (e.g. smelling and tasting) and ‘partially available senses’ where the viewer may have access to some of the sensory experience (e.g. hearing and seeing). As we shall see, this results in differences in the reviewers’ gestural and verbal accounts of the senses and how they epistemically position the viewer.

There are no clear guidelines regarding anonymisation in the analysis of publicly available data. I follow the examples of other work on YouTube where data are not anonymised (Bhatia, 2018; Jorge et al., 2018) because the facial expressions are relevant to the analysis and because the public availability of the videos means readers can visit the original data sources. However, non-anonymisation presents dilemmas for researchers as we re-contextualise content without users’ consent. While the risks of harm may be minimal, recontextualisation nonetheless raises questions about our relationship as scholars to participants and the unsolicited ‘parasitic’ nature of scholarly work (Gibson and Roca-Cuberes, 2019). While it is beyond the remit of this article to discuss such issues in depth, I hope that my analysis can inform reflection on these issues in the wider community.

The analysis pays particular attention to the intersection of talk and gesture in the production of reviews. Gestures are treated here as ‘improvised, hybrid devices for rendering aspects of the world visible and intelligible’ (Haviland, 2012: 228). A key focus of my analysis is on the use of gestures as communicative resource in the articulation of sensorial action. I draw on McNeill’s (1996) categorisation of five gesture types: iconic gestures (representing objects); metaphoric gestures (representing abstract concepts); deictic gestures (used for pointing); beats (as rhythmic movements for organising attention); and ‘butterworths’ (used in place of gestures). I also use elements of Streeck’s (2009) gestural framework which builds on a distinction between gestures that are used to describe the world and those that are used to find something about it (Haviland, 2012) (See Table 2 for an overview of Streeck’s framework).

OPEN IN VIEWER

Table 2.

Streeck’s (2009) framework of gestural analysis.

Principle	Description	Example
Making sense of the world at hand	Using the hands to show aspects of the ‘properties’ of objects in the world to others	Displaying the shape of an object by tracing hands along it
Disclosing the world within sight	Using gestures such as pointing to ‘order’ the world (for others) that is not reachable with the hands	Pointing or gesturing to an object that is out of reach
Depiction	Using gestures to represent an object relevant to the interaction (similar to McNeill’s (1996) ‘iconic’ gestures)	Holding hand to the ear to represent a phone
Displaying communicative action	Used with pragmatic functions related to talk, such as giving emphasis, punctuation and so on (Similar to McNeill’s (1996) ‘Beats’)	Using head nods or hand movements to emphasise particular words
Ordering and mediating transactions	Using gestures to ‘order the world’, like a conductor might	Using a ‘beckoning gesture’ to guide a reversing car into a parking space

Critical to this study is the idea that, like spoken language, gestures can be employed by someone to show others what they are experiencing or to facilitate the sensorial engagement of others – that is, to make others experience the world in particular ways, drawing attention to specific features of the world. Licoppe (2017) shows how gestures are used as resources to facilitate the sensorial engagement of others by analysing the construction of joint attention in mediated environments through ‘showing sequences’, which make particular aspects of one person’s context visible and interactionally relevant. This article aims to expand on this analysis by exploring how other senses are shared and made accountable through communicative actions.

Analysis

This discussion is divided into two sections: (1) reviews that deal with senses that are not available to the viewer (smell, taste and touch); (2) reviews that deal with senses that are partially available to the viewer (sound and vision).

Discussion

At the start of this article, I showed that existing work on YouTube review practices has focussed on how neoliberal discourse inform rhetorical strategies of reviewers in mediating distinctive demands such as ‘authenticity’, ‘being entertaining’ and ‘doing selling’. While these distinctive discursive modes may be important, researchers to date have not paid attention to more ‘mundane’ but no less critical issue related to the production of product descriptions. I have argued here that interactionist approaches to the senses help to show how objects are symbolically transformed from lifeless, disembodied, decontextualised ‘things’ into enlivened and experientially rich products. In this final section, I review the interactional strategies through which this occurred in the videos and discuss the implications of the analysis for the theoretical and methodological study of YouTube product reviews.

Gestures and the sensory act

In the previous section, I used the term ‘sensory act’ to describe the performance of sense work and I showed that this took two distinct structural forms. The first form involved performing the senses for the viewer, so that they can see the reviewer ‘doing’ smelling, tasting and looking. In the examples of smelling and tasting, the reviewers performed the sensory act separately from the verbal/gestural articulation of the experiential display, while in the example of seeing, the reviewer spoke and gestured while ‘doing looking’. This difference reflects in part the difficulty of smelling/tasting while talking, but the separation of the talk from the action marked them as performative in character, which fits with previous research on the ritualistic nature of perfumery and wine tasting (Alač, 2017; Shapin, 2016). In all cases though, senses were presented as something the reviewers were experiencing and reporting to the viewer.

In the second form, the reviewer invoked the viewer as a participant in the sensory act, projecting shared epistemic relations between them. This strategy was only found in the piano review (Extract 4), where the reviewer treated the piano’s sound as intersubjectively shareable in spite of the mediated (and altered) nature of the sound. In contrast, we saw in the wine review that a sharable quality of the ‘colour of the wine’ was not produced, and instead the viewer was positioned as epistemically removed from the product. In both of its forms, the sensory acts transformed the object into something enlivened by situating it within an action sequence where sensory properties could be described/witnessed/experienced, and, in turn, ascribed as properties of the objects themselves.

Turning to the details of this communicative work, we saw that gestural actions were critical to the making of accounts about sensory experiences, with presenters using different types of gestures to draw attention to and to characterise aspects of the properties of objects. First, iconic gestures were used as metaphorical devices to enact qualities of objects that were also described verbally. This was particularly the case with the senses of smell and taste – intersubjectively unavailable senses – where the gestures performed physical enactments of verbal descriptions such as ‘shape’ and ‘direction’ (Extract 1). Second, sense making gesture achieved distinctive interactional work by attempting to display something that was not articulated verbally or not clearly tied to a description. In Extract 2, we saw how the gesture of movement in describing occurred during a pause in speech and had the character of ‘looking for’ a representation rather than ‘showing’ something already articulated.

Third, I referred to gestures that had a ‘performative character’ and that acted together with speech to show something of the products’ sensory characteristics. We saw this in Extract 3, with the reviewer’s manipulations of the mobile phone, such as rubbing a finger on a phone while describing the feeling of ‘thinness’. This involved indexical actions of connecting verbal descriptions through a kind of pointing, but also rhythmic gestures to bring awareness to particular features. Finally, we also saw the use of pragmatic gestures for the purposes of punctuating, emphasising and ordering (e.g. ‘definitely’ in Extract 4). While still implicated in the communicative act of ‘showing senses’, this type of pragmatic work is distinctive to gestural characterisations that depict or ‘enliven’ products as they form ordinary modes of ‘doing talk’ rather than for ‘showing sensory work’.

Through this analysis, we see the central role that gestures play in the articulation of sensory work. Streeck’s (2009) and McNeill’s (1996) frameworks are important resources for showing the diversity of interactional work that gestures can play, and my aim has been to build on this work by showing the uses of these gestural forms for the construction of sensorial action. Much more work is needed in this area to unpack the relationship between gestures and sense work in different contexts.

In addition to gestures, we saw other semiotic orderings used to draw attention to senses, such as by positioning objects in a particular way (e.g. placing different perfumes next to each other). Sometimes, the special organisation could be seen to hinder particular viewer experiences, such as how wine bottle obscured the wine glass in Extract 5. Although it has not been the focus of this study, such practices create an ordered visual field that is evidently an important component of how sensorial characteristics are structured, and I suggest that this is a useful area for further investigation in sensory scholarship.

Implications for the study of selling discourses

As a gloss, we might say that ‘selling’ and ‘endorsing’ both involve ‘talking up’ products and making them desirable by, for example, revealing particular properties of them, their utility or purpose and why they are better than other products. However, there are three points to emphasise about the contextually specific features of such selling.

First, from the analysis, it is clear that all of the repertoires of description and enlivening need to be positioned within established and emergent review conventions. While this article does not aim to make generalised claims about review practices, it is nonetheless evident that there were commonalities across the different genres of videos. A repeated feature of mobile phone reviews is that they use a camera frame that shows the reviewers’ hands manipulating a phone without a shot of their face. Wine reviews typically involved the performance of established wine tasting rituals in the context of a wine review. Both wine and perfume reviews used the same sets of enshrined descriptors to characterise the products. These examples show that communicating sensory experience is embedded in embodied cultural routines (Crossley, 1995; Merleau-Ponty, 1962) that are critical to understanding how reviews are structured.

Second, I suggest that understanding the more localised conventions of YouTube review practice are also critical to the analysis of the performance and negotiation of identity. The small amount of existing research on YouTube reviews has pointed to striking similarities between beauty vloggers (Bhatia, 2018; Jorge et al., 2018), but very little is known about other genres of YouTube reviews. The detailed analytic description of the orders of cultural practice is necessary to describe the interpretive frameworks that reviewers work within, and these are potentially much more expansive than the quite abstract political framings of neoliberalism.

Third, the intersubjective positionings of the viewer in reviews radically transformed the relation between the viewer and the product itself and it has clear implications for the theorisation of identity formation in YouTube vloggers. For example, the relation created between reviewer, viewer and product is central to the types of performance work implicated in ‘doing reviews’, and the types of relationships that reviewers can establish between themselves and viewers. So far, this aspect has been overlooked, largely, I suggest, because of the restrictive focus that has been placed on neoliberal discourse at the expense of a wider analysis of interactional practice.

Future work

This article has not focussed on the production of evaluations as a conversational action, but on the description of products themselves. However, it became clear through the analysis that there were remarkable similarities in the sequential placement of evaluations as emergent from descriptions. Conversation Analysis has a long held interest in evaluation in terms of the production of accounts around assessment (Goodwin and Goodwin, 1987; Heritage and Stivers, 1999), how people align in the production of making assessments (Pomerantz, 1984) and, in sales contexts, the production of customer affiliation following assessments by sellers (Clark et al., 2003). I suggest that online sales environments are an important area for developing this interest further. Similarly, interactionists’ interests in video-mediated communication can be expanded beyond the focus on identifying objects (Licoppe, 2017) to look at how gestural and verbal actions are used to communicate about the multisensorial features of material contexts.

The approach taken in this article involved using everyday sense categories as an analytic frame to explore how sense work is achieved. In much of the data it is clear that the participants themselves are concerned with exactly these sense categories and with finding ways to make them intersubjectively available. In the wine review, for instance, the reviewer described the senses of ‘vision’, ‘taste’, ‘smell’ separately, producing accounts of them as distinct categories of experience. By contrast, in the mobile phone review, while the articulation of touch and feel was evident as a communicative practice and concern, this was embedded in actions of characterising size and shape as a feature of usability. I suggest that a very important line of research will be to explore further how the senses become relevant to mundane actions, where dealing with the senses is an emergent concern, rather than the purpose of the action. A related concern is the exploration of how sensorial experience of phenomena is integrated into other kinds of accounting practices, including evaluations.

I noted earlier that a common starting point for studies of the senses has been an observation that conventional sense categories are problematic as they operate as culturally specific articulations of sense that may misrepresent people’s lived phenomenological experience (Gibson and vom Lehn, 2019). While this critique is useful for illustrating the cultural relativity of sense language and experience, in this article, my concern has been with how people use conventional language (including sense categories) to bring about accounts of their experiences and to order other people’s experiences. The strategies described above give some indication of how everyday sense work is managed, but this emerging interest needs much more work to explore empirically the ways that people handle sense work as an everyday activity.