The Networked Camera at Work: Why Every Self-portrait Is Not a Selfie, but Every Selfie is a Photograph

May 2, 2016
Author Alise Tifentale
Published in Detour

Figure 1. This is not a selfie. Robert Cornelius (1809-1893). Self-portrait. Approximate quarter plate daguerreotype. October or November, 1839. Library of Congress, American Memory collection, LC-USZC4-5001 DLC.

Recently an artist friend claimed in a conversation that he thinks that people, including himself, have made selfies all the time, even before the appearance of social media and smartphones. He said he used the word “selfie” just as a shorter version of “self-portrait.” Indeed, “selfie” is closely related to the concept of “self-portrait,” but it is more than that. This article focuses on the role of technologies in defining and understanding the selfie. While there is danger of slipping into oversimplified technological determinism, we have to admit that the role of technologies in visual culture, and especially photography, is often underestimated. Could phenomena like the selfie really be just a byproduct of the advancement and accessibility of digital image-making and image-sharing technologies? Or could it be vice versa—new and emerging photographic practices shaping the design and features of hardware and apps, such as the introduction of the second camera in smartphones and appearance of Instagram and other image-sharing platforms?

What’s in the name of the selfie?

Looking back into the history of photography, the cheap and easy to use Kodak Brownie cameras around 1900 gave rise to popular and amateur photography, introduced the snapshot, and established a tradition of family photograph albums. Similarly, around 2010 we saw the rise of a new kind of image-making device, the smartphone with a built-in camera and wireless connection to the Internet. Availability of such devices in the mass market was followed by a formation of new sub-genres of popular photography, such as the selfie.

This, however, could happen only because a demand or desire for such technological innovations had already been articulated in society. Thus also the appearance of the selfie as a new sub-genre of popular photography is historically time-specific: it could emerge only in a moment when several technologies reached a certain level of development and accessibility and when a “burning” human desire had emerged, referring to Geoffrey Batchen’s highly influential book.[1] Even though many photographers—well-known artists and enthusiastic hobbyists alike—have made self-portraits since the very early days of photography, scholars have confirmed that “self-portraits did not become a mundane practice until the digital camera converged with the mobile phone.”[2] Furthermore, the convergence of the camera with the smartphone is not all that is needed for a selfie—there has to be a human desire to make such picture and—equally important—to share it with one’s peers. According to the definition by the Oxford Dictionaries, a selfie is “a photograph that one has taken of oneself, typically one taken with a smartphone or webcam and shared via social media.”[3]


Figure 2. This is not a selfie. Zenta Dzividzinska (1944-2011). Untitled (self-portrait). 1966. Scan from 35mm black-and-white film negative.

This definition neatly sums up all three key activities that are essential for the selfie: taking a photographic image of oneself, using a camera on one’s smartphone, and sharing this image on social media networks. While other scholars have introduced the term “networked image,” I would like to suggest a slightly different term that shifts the focus more toward the apparatus that produces the image: the networked camera.

The networked camera is a curious hybrid: an image-making, image-sharing, and image-viewing device whose necessary features include hardware such as an easy to use smartphone with a built-in camera, the availability of a wireless Internet connection, the existence of online image-sharing platforms, and the corresponding software, the ‘invisible hand’ that drives the devices and service platforms. This combination facilitates a streamlined production, dissemination, and consumption of visual information.

The concept of the networked camera helps to understand the selfie as a hybrid phenomenon that merges the aesthetics of photographic self-portraiture with the social functions of online interpersonal communication. Just like the networked camera is more than only a new type of camera, the selfie is more than an image. Although the selfie is reminiscent of traditional photographic self-portraiture, its other essential attributes include metadata, consisting of several layers: automatically generated data (like geo-tags and time stamps), data added by the user (hashtags), and data added by other users (comments).

Another, no less important attribute of the selfie is the instantaneous dissemination of the image via Instagram or similar social networks that makes the selfie significantly different from its earlier photographic precursors.[4] As Sonja Vivienne and Jean Burgess have observed, “much more important than digital photography’s influence on the practice of taking photographs, then, are the ways in which the web has changed how and what it means to share photographs.”[5] Agreeing on this, we could argue that the selfie contributes to the current condition of the globalized and completely born-digital visual culture, described by Lev Manovich as “softwarization” and “the new global aesthetics” that “celebrates media hybridity and uses it to engineer emotional reactions, drive narratives, and shape user experiences.”[6]


Figure 3. This is a selfie. Alise Tifentale. Untitled. January 2016. Screenshot of Instagram app on a smartphone.

These considerations can partly serve as an answer to those who tend to apply the term “selfie” retroactively to photographic self-portraits made before c. 2010. While there are lots of self-portraits in the history of photography that look seemingly similar to selfies—self-portraits in mirrors, self-portraits made while holding the camera in one’s extended arm, etc.—these images are not selfies because they are not products of the networked camera, they were not made with a camera on one’s smartphone and were not shared on social media networks. As simple as that. (Fig. 1, 2, and 3)

Already before Instagram and the selfie, some scholars had noted the dualism of online image-sharing practices—the coexistence of their aesthetic and social functions—and had observed that the emphasis in analysis most often tends to be put on the “social life of the networked image”, while overlooking the image itself.[7] Social sciences and media studies have provided a solid theoretical and methodological basis for thinking about identity construction and performance through photography in social network sites.[8] As we speak, a growing number of articles are being published about the selfie from various perspectives. An important body of most recent scholarship is collected in a special section of the International Journal of Communication, edited by Theresa M. Senft and Nancy K. Baym in 2015.[9] Some interesting insights have appeared, for example, in journals such as Consumption Markets & Culture.[10] There is also a continuous flow of publications that tend to criticize the genre of the selfie as such. A typical example is “Selfie Culture in the Age of Corporate and State Surveillance” by Henry A. Giroux, where a wide range of concepts is mobilized to discredit the selfie: “mainstream corporate selfie culture,” “pathological,” “celebrity-fed stupidity” and “insufferable idiocy,” as well as the often-repeated reference to “narcissism” without any further elaboration on how Freudian psychoanalysis could help us understand today’s popular photography.[11] While all these contribute to the debate, what interests me are the mechanisms at work in the construction of the selfie as a new sub-genre of popular photography.

The distant reading of the selfie

As an example of such inquiry I would like to present the research project Selfiecity (, 2014) that was led by Lev Manovich and his research lab Software Studies Initiative based in The Graduate Center, City University of New York. The research team, apart from myself, included Dominikus Baur, Jay Chow, Daniel Goddemeyer, Nadav Hochman, Moritz Stefaner, and Mehrdad Yazdani. The object of the research was a dataset of 3,200 selfies posted to Instagram during one week in 2013 from five global cities: Bangkok, Berlin, Moscow, New York, and Sao Paulo (Fig. 4). Recently, a new set of selfies posted during one week in September 2015 was added and analyzed in Selfiecity London (, 2015) that was commissioned for the exhibition Big Bang Data (Somerset House, London, December 3, 2015–March 20, 2016).


Figure 4. Imageplot: selfies from New York, arranged in a grid. Part of the research project Selfiecity, 2014. Image courtesy

Selfiecity and Selfiecity London offers a comparative reading of selfies posted from different cities, and our leading research questions were focused on finding out whether there is a significant cultural difference at play or “all the selfies everywhere are the same.” For this inquiry, various computational analysis methods (such as software-driven face recognition and use of custom-made data visualization tools) were applied as well as formal aesthetic analysis of each individual image. Computational methods were used to analyze characteristics such as pose (for example, looking up/down, left/right), facial expression, and guess mood. Research tools included media visualizations, imageplots, blended video montages, and a custom-made interactive web application Selfiexploratory (Fig. 5). This application as well as examples of other methods are available online at We found that selfies indeed differ from city to city, while formally they belong to the same genre with its own aesthetic conventions. These and other findings of the project, as well as the nuances of research methodology have already been extensively discussed elsewhere.[12] What I would like to elaborate here is some considerations for future research.


Figure 5. Screenshot of the custom-made web application Selfiexploratory that offers tools to explore the Selfiecity dataset. Part of the research project Selfiecity, 2014. Image courtesy

The starting point of research projects like Selfiecity is inevitably a dataset. First, team members agree on a definition of their object of study and then collect a sufficient number of examples to analyze. In the case of Selfiecity, images were downloaded from Instagram API (application programming interface) and then manually filtered to select selfies. We did not rely on a perhaps easier method of searching for images that have the hashtag “selfie” because not all Instagram users actually tag their selfies as “selfies” and not all images tagged with “selfie” are selfies, and also because of the multiple languages involved in our selected cities across all continents. After the team has constructed this “clean” dataset of selfies, the analysis starts.

However, such a method partly undermines the earlier agreement that selfie is a product of the networked camera, which is simultaneously an image-making, image-sharing, and image-viewing device. Even if we acknowledge the specifics of making (smartphone photography) and sharing (Instagram), we cannot fully grasp the third element—viewing. Such a “clean”—carefully edited and curated—dataset of selfies with equal number of examples representing each city is definitely not the most typical way that these images are encountered by Instagram users on an everyday basis. The selfies we study are samples in artificially constructed sets which none of the Instagram users have ever experienced directly. This approach is what Franco Moretti has called “distant reading.”[13] Such distant reading—the way a dataset of selfies in Selfiexploratory is being viewed and perceived—is radically different from the so-called close reading, a careful and attentive inspection of one single item at a time. The way that each selfie is viewed and perceived in the time of its appearance on Instagram, perhaps, is more reminiscent of close reading.[14] At least, such images are usually inspected one by one. While we look at the aggregate of thousands of images, we also have to keep in mind the original environment where these images are being viewed by their intended audience. In our global dataset, each image is extracted from its original context and removed from its initial mode of appearance (on a smartphone screen), it is disconnected from the larger body of images and textual input that makes people’s Instagram accounts a communication tool, meaningful for their followers. The distant reading can be complemented by the close reading, and that would include also a closer look at the ways that selfies are viewed by their intended audience.

Selfie and its natural habitat

First of all, selfies most naturally appear on the Instagram screen on one’s handheld device, most typically a smartphone. That points to further consideration of the most typical times and places of viewing Instagram content as well as the mode of viewing (individual, intimate, and close-up). Second, there are only a few most common layouts available for a typical Instagram user to view the content uploaded to the app.

For example, the “home” screen presents a real-time flow of image contributions by the people whom one follows (Fig. 6). Each image takes the full width of the screen, and scrolling up will reveal the author’s caption and assigned hashtags as well as other users’ comments underneath, followed by next image. In this flow, selfies—just like any other content posted by the people one follows—are viewed separately, one by one, but as a part of a stream of other images and videos. The other most common view is a grid of thumbnail-size square images—gallery view— which one can access by clicking on a user’s name (Fig. 7). Fifteen thumbnails would be the average number of images that a user can see on a single screen. In this grid, selfies also appear in context with other content that the particular user has posted. Finally, there is also a “search” screen that offers a view that is the closest to an edited and curated database (Fig. 8). Search results appear in a grid of an average fifteen square thumbnail images per screen, like in a user’s gallery view. This interface provides searching only for a single parameter at a time (“top, “people, “tags,” and “places” are the options).

In addition, in all three views images appear in chronological order of their posting. If some users in their accounts are constructing a time-based narrative about their experiences, the temporal aspect is completely lost when individual selfies are extracted and compared. Furthermore, the flow on Instagram is live, and interaction from the person viewing images is welcome and encouraged—one can click the heart symbol (“like” the image), post a comment, forward and re-post an image, or take and post their own image at any time. Activities that were clearly separated in earlier moments in the history of photography now converge in a device which is at the same time an image-making, image-sharing, and image-viewing machine. This aspect is lost when images are singled out and removed from their natural media environment.

Such questions are nothing new to art history and, to a lesser degree, also to the history of photography. When dealing with born-digital photographic images, however, it seems that it is even easier to forget the medium and to overlook the fact that, before everything else, each selfie is a photograph. A photograph that has its own historically specific mode of making, sharing, and viewing.


Figure 6. Regular mode of viewing images on Instagram. Screenshot of Instagram app on a smartphone. February 2016.


Figure 7. Gallery view of Instagram user Alise Tifentale’s account. Screenshot of Instagram app on a smartphone. February 2016.


Figure 8. Search results screen on Instagram (search by the tag #selfie). Screenshot of Instagram app on a smartphone. February 2016.


The issues addressed in this article are based on my involvement in the research project Selfiecity (2014). I wish to thank Lev Manovich from whose guidance and generous support I have benefited while working on this article. I am especially grateful to Hon Sun Lam for the inspiration and encouragement.

[1] Geoffrey Batchen, Burning with Desire: The Conception of Photography (Cambridge, Mass: The MIT Press, 1997).

[2]Marika Lüders, Lin Prøitz, and Terje Rasmussen, “Emerging Personal Media Genres,” New Media & Society 12, No. 6 (2010): 959.

[3] Oxford Dictionaries, “selfie.” Oxford University Press. Available at Accessed February 19, 2016.

[4] See Kandice Rawlings, “Selfies and the History of Self-Portrait Photography,” Oxford University Press Blog, November 21, 2013. Available at Accessed February 19, 2016.

[5] Sonja Vivienne and Jean Burgess, “The Remediation of the Personal Photograph and the Politics of Self-Representation in Digital Storytelling,” Journal of Material Culture 18, No. 3 (2013), 281. Emphasis in original.

[6]Lev Manovich, Software Takes Command (New York, London: Bloomsbury, 2013), 179.

[7] See, for example, Daniel Rubinstein and Katrina Sluis, “A Life More Photographic, Mapping the Networked Image,” Photographies 1, No. 1 (2008), 9-28.

[8] See, for example, Zizi Papacharissi, ed., A Networked Self: Identity, Community and Culture on Social Network Sites (New York: Routledge, 2011).

[9] See the editors’ introduction, Theresa M. Senft and Nancy K. Baym, “What Does the Selfie Say? Investigating a global Phenomenon,” International Journal of Communication 9 (2015), 1588–1606. This and all other articles in this section are available at Accessed February 19, 2016.

[10] Derek Conrad Murray, “Notes To Self: The Visual Culture Of Selfies in The Age of Social Media,” Consumption Markets & Culture 18, No. 6 (2015), 490-516.

[11]Henry A. Giroux, “Selfie Culture in the Age of Corporate and State Surveillance,” Third Text 29, No. 3 (2015): 155-164.

[12] See, for example, Alise Tifentale and Lev Manovich, “Selfiecity: Exploring Photography and Self-Fashioning in Social Media” in David M. Berry and Michael Dieter, eds., Postdigital Aesthetics: Art, Computation and Design (London: Palgrave Macmillan, 2015), 109-122; Alise Tifentale, “Art of the Masses: From Kodak Brownie to Instagram,” Networking Knowledge 8, No. 6 (2015): 1-16; Alise Tifentale and Lev Manovich, “Competitive Photography and the Presentation of the Self” in by Jens Ruchatz, Sabine Wirth, and Julia Eckel, eds., #SELFIE–Imag(in)ing the Self in Digital Media (Marburg, forthcoming 2016).

[13]Franco Moretti, Graphs, Maps, Trees: Abstract Models for Literary History (New York: Verso, 2005), 1.

[14] The coexistence of the distant reading with the close reading is at the heart of an ongoing discussion in the field of digital humanities. See, for example, David M. Berry, ed., Understanding Digital Humanities (Basingstoke: Palgrave Macmillan, 2012); Anne Burdick, Johanna Drucker, Peter Lunenfeld, Todd Samuel Presner, and Jeffrey T. Schnapp, Digital_Humanities (Cambridge, Mass: The MIT Press, 2012); Ted Underwood, Why Literary Periods Mattered: Historical Contrast and the Prestige of English Studies (Stanford, California: Stanford University Press, 2013).

This text was originally published in the catalogue of Riga Photography Biennial 2016.