PROOF COVER SHEET

Author(s): Ilya Utekhin

Article Title: Small data first: pictures from Instagram as an ethnographic source
Article No: RRJC1327328

Enclosures: 1) Query sheet

2) Article proofs

Dear Author,

1. Please check these proofs carefully. It is the responsibility of the corresponding author to check
these and approve or amend them. A second proof is not normally provided. Taylor & Francis
cannot be held responsible for uncorrected errors, even if introduced during the production process.
Once your corrections have been added to the article, it will be considered ready for publication.

Please limit changes at this stage to the correction of errors. You should not make trivial changes,
improve prose style, add new material, or delete existing material at this stage. You may be charged
if your corrections are excessive (we would not expect corrections to exceed 30 changes).

For detailed guidance on how to check your proofs, please paste this address into a new browser
window: http://journalauthors.tandf.co.uk/production/checkingproofs.asp

Your PDF proof file has been enabled so that you can comment on the proof directly using Adobe
Acrobat. If you wish to do this, please save the file to your hard disk first. For further information
on marking corrections using Acrobat, please paste this address into a new browser window: http://
journalauthors.tandf.co.uk/production/acrobat.asp

2. Please review the table of contributors below and confirm that the first and last names are
structured correctly and that the authors are listed in the correct order of contribution. This
check is to ensure that your name will appear correctly online and when the article is indexed.

Sequence Prefix Given name(s) Surname Suffix

1 Ilya Utekhin
Queries are marked in the margins of the proofs, and you can also click the hyperlinks below.
Content changes made during copy-editing are shown as tracked changes. Inserted text is in red font
and revisions have a red indicator . Changes can also be viewed using the list comments function.
To correct the proofs, you should insert or delete text following the instructions below, but do not
add comments to the existing tracked changes.

AUTHOR QUERIES

General points:
1. Permissions: You have warranted that you have secured the necessary written permission from
the appropriate copyright owner for the reproduction of any text, illustration, or other material in
your article. Please see http://journalauthors.tandf.co.uk/permissions/usingThirdPartyMaterial.asp.
2. Third-party content: If there is third-party content in your article, please check that the
rightsholder details for re-use are shown correctly.
3. Affiliation: The corresponding author is responsible for ensuring that address and email details
are correct for all the co-authors. Affiliations given in the article should be the affiliation at the
time the research was conducted. Please see http://journalauthors.tandf.co.uk/preparation/
writing.asp.
4. Funding: Was your research for this article funded by a funding agency? If so, please insert
‘This work was supported by <insert the name of the funding agency in full>’, followed by the
grant number in square brackets ‘[grant number xxxx]’.
5. Supplemental data and underlying research materials: Do you wish to include the location
of the underlying research materials (e.g. data, samples or models) for your article? If so,
please insert this sentence before the reference section: ‘The underlying research materials for
this article can be accessed at <full link>/ description of location [author to complete]’. If your
article includes supplemental data, the link will also be provided in this paragraph. See
<http://journalauthors.tandf.co.uk/preparation/multimedia.asp> for further explanation of
supplemental data and underlying research materials.
6. The CrossRef database (www.crossref.org/) has been used to validate the references. Changes
resulting from mismatches are tracked in red font.

QUERY NO. QUERY DETAILS

AQ1 Please provide a short biography of the author.
AQ2 Please provide the missing department detail for the affiliation.
AQ3 Figures 1–9 were not cited in the text. So, citations have been inserted. Please provide
corrections if these are inaccurate.
AQ4 ‘interpretations that allow for read the visualizations’ – Please check this expression
for clarity since the meaning is not clear.
AQ5 The year for “Daer et al., 2014” has been changed to “Daer et al., 2015” to match the
entry in the references list. Please provide revisions if this is incorrect.
AQ6 The disclosure statement has been inserted. Please correct if this is inaccurate.
AQ7 The CrossRef database (www.crossref.org/) has been used to validate the references.
Mismatches between the original manuscript and CrossRef are tracked in red font.
QUERY NO. QUERY DETAILS

Please provide a revision if the change is incorrect. Do not comment on correct

changes.
AQ8 Please provide the missing publisher and city for the reference “Boitsova, 2013”.
AQ9 Please provide the missing page range, publisher and city for the reference “Kontareva
et al., in press”.
How to make corrections to your proofs using Adobe Acrobat/Reader

Taylor & Francis offers you a choice of options to help you make corrections to your proofs. Your
PDF proof file has been enabled so that you can edit the proof directly using Adobe Acrobat/Reader.
This is the simplest and best way for you to ensure that your corrections will be incorporated. If you
wish to do this, please follow these instructions:

1. Save the file to your hard disk.

2. Check which version of Adobe Acrobat/Reader you have on your computer. You can do this by
clicking on the “Help” tab, and then “About”.

If Adobe Reader is not installed, you can get the latest version free from http://get.adobe.com/reader/.

3. If you have Adobe Acrobat/Reader 10 or a later version, click on the “Comment” link at the
right-hand side to view the Comments pane.

4. You can then select any text and mark it up for deletion or replacement, or insert new text as
needed. Please note that these will clearly be displayed in the Comments pane and secondary
annotation is not needed to draw attention to your corrections. If you need to include new sections
of text, it is also possible to add a comment to the proofs. To do this, use the Sticky Note tool in the
task bar. Please also see our FAQs here: http://journalauthors.tandf.co.uk/production/index.asp.

5. Make sure that you save the file when you close the document before uploading it to CATS using
the “Upload File” button on the online correction form. If you have more than one file, please zip
them together and then upload the zip file.

If you prefer, you can make your corrections using the CATS online correction form.

Troubleshooting

Acrobat help: http://helpx.adobe.com/acrobat.html

Reader help: http://helpx.adobe.com/reader.html

Please note that full user guides for earlier versions of these programs are available from the Adobe
Help pages by clicking on the link “Previous versions” under the “Help and tutorials” heading from
the relevant link above. Commenting functionality is available from Adobe Reader 8.0 onwards and
from Adobe Acrobat 7.0 onwards.

Firefox users: Firefox’s inbuilt PDF Viewer is set to the default; please see the following for
instructions on how to use this and download the PDF to your hard drive: http://support.mozilla.org/
en-US/kb/view-pdf-files-firefox-without-downloading-them#w_using-a-pdf-reader-plugin
 RRJC1327328 Techset Composition India (P) Ltd., Bangalore and Chennai, India 5/11/2017

RUSSIAN JOURNAL OF COMMUNICATION, 2017

https://doi.org/10.1080/19409419.2017.1327328

5 Small data first: pictures from Instagram as an ethnographic

source
Ilya Utekhin
AQ2 European University at St. Petersburg, Petersburg, Russian Federation
10 ¶
ABSTRACT ARTICLE HISTORY
Research projects that employ big data and data visualization Received 27 April 2017
approach for analysis of users’ posts on Instagram deal with Accepted 2 May 2017
materials devoid of contextual information about the
KEYWORDS
circumstances of taking photos and posting, but instead bear Visual culture; Instagram;
15
traces of users’ activities in the form of meta-data such as photographs; hashtags; social
location, time and date. However, in order to be interpreted, networking; digital
patterns revealed in big data require reference to the ethnography
characteristics of places and social practices. More traditional
studies of vernacular photography and its uses, as well as studies
of social networking and mobile technologies within digital
20 anthropology, are able to provide rich layers of meaning to users’
Coll:

content from Instagram that can also be considered in the

perspective of visual anthropology. In order to check what
features and parameters can be relevant for interpretation of data
from Instagram, a ‘small data’ approach has been used for manual
QA:

collection, annotation and analysis of a corpus of photographs

that have to do with a Ukrainian town.
25
CE: KRB

Users’ content posted in social networking sites has attracted significant amount of
researchers’ attention and has proven to be a new type of data source. Its novelty as a
30 data source consists of a combination of qualities. To start with, Instagram pictures are
naturally occurring traces of users’ activities; however, they are not, in any way, provoked
or affected by a researcher influence. On Instagram, people represent their social world
and themselves, express their opinions, and practice social networking – all these, to
some extent, are observable. Moreover, in tagging and captioning, users employ their
35 own wording and thus provide access to vernacular categories.
Before the advent of the new medium for visual communication provided by the com-
bination of digital photography, the Internet, smartphone cameras and social networking
on the Internet, amateur photography was something different: less ubiquitous, requiring
relatively more effort, more tangible and material, not so immersed in the texture of every-
40 day life, and more difficult to access. The understanding that practices of vernacular visual
culture is a promising area of study is a relatively recent phenomenon within the visual
anthropology field. Classical textbooks on visual anthropology – Collier and Collier
(1986), Hockings (2003/1974), Banks and Morphy (1997), to name a few – do not
include chapters on vernacular photography, and the study of family snapshots, for
45
CONTACT Ilya Utekhin utekhin@yandex.ru; ilya.utekhin@gmail.com
© 2017 Informa UK Limited, trading as Taylor & Francis Group
 2 I. UTEKHIN

example, Chalfen (1987) used to be conducted out of mainstream topics in visual studies
as discipline. The analysis of pre-digital amateur snapshot practices, in these studies, have
turned out to be useful when we look to Instagram materials, although it cannot cast much
light over the uses of digital images on social networking sites. The question arises
50 whether in Instagram we find a new domain of reality, made publicly available for
observation.
Instagram as a technology when used by researchers to collect visual materials might
eventually have a great effect on the what and the how of the scholar’s inquiry. Previous
technologies that brought about radically new approaches to anthropological fieldwork
55 were photography and audio recording. Photography made it possible to record exact
details of situations which the researcher did not understand, and provided grounds
and means for new forms of interaction and rapport with informants. Audio recording,
for example, in folklore studies, among other things, shifted the focus from texts alone
to also embracing their context and the way the texts are performed. The use of visual
60 methods and audio recording led in the second half of the twentieth century to the emer-
gence of visual anthropology and partial re-shaping of the whole discipline. However, the
new domains that the camera and the recorder enabled scholars to study had already
existed before, independently of that technology. In case of Instagram and other social
networks on the Internet, we are witnessing new social practices that challenge traditional
65 visual studies and visual anthropology.
Pioneering works in cultural analytics by Lev Manovich and his school employ visualiza-
tions of big data for cultural analysis. Their approach ranges from studying dynamics of
purely formal patterns while considering the photos only as assemblages of color pixel
to regarding some aspects of their content that can be measured and compared quanti-
70 tatively, and so can reveal patterns of societal systems, as in Phototrails project. A variety of
spatial, temporal, and social patterns can be found within a particular city, as evident from
case studies such as Hochman and Manovich (2013), which were based on 200,000 photos
uploaded to Instagram from Tel-Aviv over three months. The real-time dynamics of a his-
torical event was analyzed through studying visual materials from social networks using
75 geolocation data, as in the case of the 2014 civil uprising in Kiev, Ukraine, studied by a
team led by Manovich.1 This project also involved manual tagging of data, that is, some
elements of small data approach.
In some respects, big data approach reminds us of how, for example, data on court
decisions can be subject to quantitative analysis, and thus correlations can be shown
80 between the outcome and the social status of the defendant (or just the distance to
Moscow). The Institute for the Rule of Law is a research center in European University at
St. Petersburg that widely practices that sort of analysis. For instance, they can show
that when a driver hits a pedestrian and is brought to the court, the outcome depends
on the price of the driver’s car: the more expensive the car, the more possible the
85 victim–defendant reconciliation and the less probable a real prison sentence, with the sus-
pended prison sentence probability also dropping significantly (Kurmangaliyeva & Skou-
garevskiy, 2017).
Big data methods, in the studies by Manovich and his followers, reveal patterns without
deleting source materials or substituting them with abstractions. Thus, the visualization of
90 all Instagram pictures selected by hashtags related with the Hermitage Museum in
St. Petersburg, as in a recent study by Kontareva, Rudenko, and Ryzhakov (in press)
 RUSSIAN JOURNAL OF COMMUNICATION 3

makes up a huge canvas containing real pictures that can be accessed in their original
form by simply clicking on them. This is an important feature that might seem part of
the tendency to which audiovisual methods contributed in social science: instead of pre-
senting only the findings in terms of researcher’s categories, contemporary ethnographies
95 tend to include more and more of subjects’ own voice, with extensive quotes, links to
videos and, generally, providing the readers more information of the context of the
materials quoted, if not an access to source materials. All these make it possible for the
reader to trace the researcher’s way of producing new knowledge. However, the actual
difference between big data and ethnography is much more important, and they deal
100 with the fact that while big data are digital traces of human activity detached from
their context, ethnography relies on deep and personal immersion of the researcher in
a social environment in order to get access to practices and their contexts, to people’s
values, attitudes, and meanings. Is there a way to fill the gap between patterns revealed
in big data and meanings that groups of people ascribe to elements of their lifeworlds?
105 In the study by Kontareva et al. (2017), what initially became visible to scholars, in the
huge canvas that has brought together the Hermitage-related pictures in chronological
order, is the pictures’ brightness value changing across time. This might have to do
with improvements in smartphone camera technology and increased sensibility that
enabled users to take brighter pictures in low lighting conditions, or changing the aes-
110 thetic tastes of the masses, or something else. However, we also see a contrasting dark
line containing hundreds of pictures on the canvas. It turns out to be a public event
with projections mapped onto the buildings celebrated on the Palace Square, right
before the Hermitage. What we get from this discovery is the knowledge that the light
show amassed an exceptional number of Instagram posts, and that a significant part of
115 them were tagged #hermitage. Much more informative are the ways in which people
interact with the exhibits, which we can see from a manual analysis of certain sample of
the photos from the corpus of the Hermitage data.
Patterns that can be seen in data visualizations do not lead automatically to discovery
of new knowledge, as they require interpretation. The observations to be interpreted in
120 the projects consist in that the photos concentrate in certain areas of the city and thus
selectively represent the city, or that certain types of content in photos (e.g. meals) or
photos tagged with certain hashtag tend to appear in certain places (e.g. where restau-
rants and cafés are concentrated). Much of it appears rather trivial, unless dynamic pat-
terns of posting activity are considered across seasons, days of the week and time of
125 the day.2 For interpretation to become richer, there is a need for some qualitative data
that show how the patterns observed in time and space are linked with practices of differ-
ent groups of people. To make sense of quantitative data on selfies, Alise Tifentale (2015)
started from reviewing literature on the meaning of selfie. However, to my knowledge, no
ethnography has been dedicated, so far, to the practice of taking selfies and uses of them
130 by a group of people, though elements of such ethnographies are part of the projects by
research group directed by Daniel Miller et al. (2016).
In the studies where mass users’ data are analyzed, the city is the protagonist of the
story told by researcher, whereas the people who post on Instagram, their motivations
and practices are just pixels in the large picture. However, digital anthropology as pre-
135 sented in the publications of Miller and his group focuses on people and their practices
and thus helps to bridge the gap mentioned above.
 4 I. UTEKHIN

People and place before and after Instagram

Looking back into pre-Instagram era, we could argue that some functions that these
days belong to Instagram photos, that is, practices that relate people and places,
140
were present in a different form in tourist photography and in leaving material
traces in the environment. Let us consider some examples from Instagram that illus-
AQ3 trate this claim (Figure 1).
¶ This split screen photo was posted by a Russian tourist in Imatra, a town in Eastern
Finland, on Vuoksa river, near the Russian border. The same user had previously posted
145
this picture in his previous already discontinued account, and the caption was as
follows (in Russian):
One more Finnish town is checked out. [emoji and flags] Small, green and comfortable town of
Imatra. [emoji] Amazing holidays! People from St. Petersburg are lucky to have nearby such
beauty. So they call it, it’s a window to Europe!
150
This lengthy caption illustrates much of what in similar captionless pictures might be left
unarticulated, but nevertheless present as communicative intent and emotional content of
the message. The author is sharing his positive feelings about his holiday and the place he
visited with everyone who is following him and gets his posts in the newsfeed. This is the
part of his self-image presented by means of Instagram photos.
155
The Imatrankoski rapids with its dry riverbed are a tourist attraction, and Valtionho-
telli Hotel that appears in one of the photos is an architectural landmark. The riverbed
is a strange sight, revealing the harsh rocky bottom. The large dam behind the bridge
was built for the hydroelectric plant. In summertime, the floodgate is opened daily for

160

165

170
Colour online, B/W in print

175

180 Figure 1. Caption provided by the user, in Russian: ‘Some more of Imatra in your newsfeed. [emoji and
flags] Beautiful town! [emoji]’.
 RUSSIAN JOURNAL OF COMMUNICATION 5

tourists for about 20 minutes, and visitors can get an idea of what the rapids used to
be like. Another Instagram photo from the same place taken by a Finnish tourist
reveals in the background not only the landscape, but also something else: walking
down by the riverbank, you can see a lot of engravings on the rocks. The graffiti
185 dated from the nineteenth century display memories of visits by ordinary people
and even monarchs: Pedro II, the king of Brazil, came over with only three companions
in 1876 (Figure 2).
These days the traces from the nineteenth century have been renovated and exhibited
to the public as a curiosity. To leave such traces in a historical site today would most prob-
190 ably be considered vandalism. Cultured people are taking pictures and post them on Insta-
gram, instead.
Within the rhetoric of the city as protagonist, it can be said that Instagram pictures are
part of the image of the place that people are producing. We keep in mind, however, that
an aggregate image of the place is a figure of speech: people share the same stereotypes
195 and the visual culture of their group, but act independently and follow their own motives,
so the images that they produce belong to them, not to the place. In her study of family
snapshots within the pre-digital era in Russia, Olga Boitsova (2013) considered not only the
content, genre, composition and style of the photos, but also how the pictures were used,
where they were kept, in which way, to whom and on what occasions they were shown,
200 how they were inscribed, organized and captioned in the albums. Some of these practices
have their digital counterparts in the functionalities provided by social networks on the
Internet.
To take a picture does not require much effort these days. However, by tagging the
picture by hand (not by tagging software), the user makes a conscious effort, and this is
205

210

215
Colour online, B/W in print

220

225
Figure 2. A Finnish tourist at Imatra rapids, with graffiti on the rocks.
 6 I. UTEKHIN

much more informative than automatically generated geolocation data. This effort might
not involve anything beyond the user’s intent to make the image easier to find – to other
people, and particularly, to the user’s followers, but also to himself/herself. At the same
time, this also might mean a contribution to an imagined – or real – dialogue whose par-
230 ticipants have used the same hashtag. That is to say, among the motives behind tagging
and captioning, there is a sort of competition in self-expression. For ethnographer, all prac-
tices of picture taking and using, posting, tagging and communicating with images could
become sources of data.
Having in mind the proliferation of photos in which the author of the Instagram post is
235 represented, it is worth indicating two types of stories that are linked to the subject of the
picture. First, the message can be a memory of an experience that the author had together
with his/her addressee or addressees. This is the ‘us’ of the picture, and ‘we’ can be shown
together in the photo. Second, it is a story about ‘me’, the author and the sender, that is, I
was there and saw this, and I address this to you who was not there with me.
240 Imagine that someone is showing to you a real album with photographs printed on
paper. The sequence of showing is determined by the design of the album and by how
the person showing the album chooses a sequence to make up a story and to link the
images. She can point at some details in the pics, refocus your attention and provide com-
ments. In Instagram, the story is told by the pictures themselves, their captions, hashtags,
245 as well as the composition, the filters used and probably the sequence of presentation if
you are watching the pictures of a particular user.

Exploring an image of a Ukrainian town on Instagram

250 All of these I had in mind when in early 2015 Alexander Shirokov from Moscow School of
Social and Economic Sciences asked me to join a study aimed at finding the ways in which
Instagram photos could be used to describe how Instagram users represented the city of
Nizhny Novgorod. I had read about Lev Manovich’s work with similar material and was
conscious of the fact that data visualization techniques and statistics would not bring
255 about significant results by themselves, apart from interpretations that allow for read
AQ4 the visualizations as tools for cultural analysis.
¶ Any study of this kind requires participation of software developers, in order to prepare
and customize the scripts that would collect data, perform statistical analysis, and visual-
ization. This is done either by using ready-made tools customized for the task, or by devel-
260 oping specialized instruments. In this case, if the researcher is versed enough in
programming, this becomes easier, but programmers still need a specification in order
to understand what functionality is required from the software product, that is, what
kind of data are supposed to be collected. This decision is crucial for the result being
informed with the ideas of the researcher about which features are relevant for the task.
265 Since it was my first experience using Instagram pictures as a source, I started from a pilot
study that would help me to formulate technical requirements for the software needed to
extract and analyze data. In November 2015, I collected by hand, using the web page source
view option of my browser, all Instagram photographs with hashtags naming a small Ukrai-
nian town, with the population of about 5000 people, where I happen to spend my summer
270 holidays. This resulted in a small corpus that I could review entirely without getting lost in
thousands of pictures and user profiles. So what if instead of 400,000 pictures we have some
 RUSSIAN JOURNAL OF COMMUNICATION 7

400? We would hardly be able to see dynamics and large-scale patterns invisibly present in
our small collection, but still we can study what kind of information is possible to read from
the data. This small dataset reflects a social environment and locations that I hopefully can
recognize and where I expect to at least partly understand the meaning of the messages
275 behind the posts. If some parameters prove relevant on this dataset, would not it be reason-
able to suppose that for a much bigger sample, the same parameters would also remain
meaningful alongside with the new opportunities open by big data approach? And is it
possible to disregard the findings of this kind of small-scale ‘textual ethnography’ approach
while interpreting big data? These questions are to remain rhetorical here, as I do not yet
280 propose any answers.
Initially moved by a very fuzzy interest in how the place is represented in Instagram pic-
tures, I wanted to find some traces of vernacular Instagram practices to understand what
types of messages are conveyed by the posts and, more broadly, what practices of Insta-
gram allow us to penetrate the meanings that the users ascribe to places.
285 Hashtags that refer to the town are more than one, and this mirrors the sociolinguistic
situation in that community: both Ukrainian and Russian languages are known to the popu-
lation, with Ukrainian being the predominant language of everyday communication, televi-
sion, business and schooling. At the same time, Russian also remains in use both in everyday
activities and as a language of popular culture. So, the following hashtags are in use:
290 #шишаки Rus /Ukr
#shishaki Rus/Ukr/Eng
#shyshaky Ukr/Eng

Russian and Ukrainian names coincide, as both languages use Cyrillic, and mean literally
295 ‘the hills’. However, its Latin transliteration – ‘Shyshaky’ – reflects Ukrainian pronunciation.
Since I was interested primarily in the practices of ordinary users of Instagram, I started
by trying to filter out commercial advertising, mostly from retail delivery services showing
the items that are sold and delivered to several towns listed as hashtags, Shishaki among
them. They do not reflect any representation of the place, showing the goods for sale
300 instead. After the filtering, I got a set of 54 pics from 28 users, corresponding to the
period of 2 years and a half between the arrival of Instagram to Shishaki in spring 2013
and the moment of collecting data in November 2015.3 I also used the Explore Location
function on Instagram that returned 333 pictures in search results, combining the
photos marked with Shishaki as location without Shishaki as a hashtag, and photos that
305 contain Shishaki in captions or comments. A number of pictures with Shishaki hashtags
contained commercial advertising that I decided I would rather not take to my corpus,
and some appeared not to have anything to do with Shishaki the town (for instance, a
young designer from Vladivostok chose this word for her account, apparently because
her family name sounds alike).
310 My attempt at filtering out commercial advertising as something irrelevant turned out to
be revealing: advertisements by small local businesses are a part of vernacular social activity,
and hence cannot be disregarded as spam. For instance, a café that was recently opened is
advertised by the owner through social networking sites. Most of the pictures posted in the
café’s Instagram account represent celebrations and parties, that is, what people typically
315 do in the café, and in this respect, they are like any other picture showing parties and
friends having good time from Explore Location set. Even the photographs of nails from
 8 I. UTEKHIN

Explore Location collection are not always just a commercial offer: as we can see from the
caption, it might be a message of gratitude to a particular manicure master.
Explore Location predictably showed more selfies of various kinds, while the tagged
picture set contained more landscapes, sites of natural beauty, pictures representing
320 environment and details of everyday life. No visualization is needed to grasp this differ-
ence: it is enough to take a look at the screenshots of Instagram search results to under-
stand that those who do not care about specifically mentioning Shishaki, by tagging with
their pictures, are mostly saying something about themselves, whereas those who did tag
the photos with #shishaki or its equivalents, are also wishing to say something about
325 Shishaki or about something belonging to Shishaki in some way (Figures 3–6).
Even though the Explore Location collection also contains landscapes, meals, flowers,
and other genres of photos, it is selfies and group portraits that noticeably dominate
here. In order to be able to use this information in a more formal way, I needed to establish
categories that characterize each picture type in terms of content, and also take into
330 account hashtags, captions, replies, date of posting, number of likes, but also data
about the source user such as the number of posts, number of followers, and user location
(if present).

335

340

345

350
Colour online, B/W in print

355

360
Figure 3. Screenshot of typical explore location results.
 RUSSIAN JOURNAL OF COMMUNICATION 9

365

370

375
Colour online, B/W in print

380

385
Figure 4. Screenshot of typical search results for pictures tagged #шишаки.

Obviously, by means of tracing sources of likes and replies, it would be possible to

explore social networks on Instagram and to identify some users’ communities even in
390
our small data sample. However, this would not allow us to find real users’ social networks
on the Internet. Nor would we be able to collect data on communication brought about by
the photographs because the pictures posted on Instagram are often reposted to VK.com
where much of the networking takes place. Nevertheless, some posts in Instagram gener-
ate followers’ replies in the comments, and it is interesting to see what in a post makes it to
395
produce communication; containing extraordinary things and arousing strong emotions is
certainly one of such features (Figure 7).
I compiled a table to bring together the information I considered relevant about the
photos and their authors, and some columns of the table were dedicated to my categor-
ization and analysis of the pictures’ content. Future algorithms will get the same job of cat-
400
egorization and tagging done faster and more efficiently than that done by human
experts, but humans are required to teach the algorithms. Amazon Mechanical Turk
service is sometimes used for similar tasks, given the fact that Artificial Intelligence,
even in its current state, can be taught to reliably recognize not just faces, but individuals,
to read textual content, not to mention the sky, water and other objects represented in the
405
photographs. However, for an ethnographer, to work with photographs and users’
 10 I. UTEKHIN

410

415

420
Colour online, B/W in print

425

430

Figure 5. Screenshot of typical search results for pictures tagged #shishaki.

Instagram accounts is a meaningful procedure in itself. Traces of real-life contexts emerge

435 as part of a whole picture. At content analysis, categories are not given in advance, but
rather established empirically while going through the data, and each new category pro-
vokes further questioning of the material, particularly, in terms of the granularity of
description.
I looked, first, if the user is in the picture, and if yes, whether it is a selfie or a picture
440 taken by another person. Then, I distinguished between different kinds of selfies, and
when the post’s author was there together with someone else, I took notice of the
gender composition of the group. I also noticed whether communicative intention was
visualized in the picture, for example, if there was a gesture addressed to us as viewers
or not. I paid attention to the setting and the background, and checked if the picture
445 was taken inside or outside, if it was a landscape or built environment, and, more generally,
whether the background was relevant for the message or, at least, legible. Along with
content analysis, I also checked whether filters, frames, or split screen compositions
were used.
Another column for my tags I called Mood, as it reflected the overall message or inten-
450 tion of the picture, and also emotional content. Sample Mood categories are: happiness,
friendship, love, beauty, fashion, psychedelic, nostalgic, cute, quiet, rest, indignation,
 RUSSIAN JOURNAL OF COMMUNICATION 11

455

460

465
Colour online, B/W in print

470

475
Figure 6. Screenshot of typical search results for pictures tagged #shyshaky.

travel, and fun. The tag set seems heterogeneous, as it brings together results of my non-
automatic sentiment analysis as well as some content and aesthetic categories, all of it
480 intended to give suitable one-word description of the main message of the post. Since
there was only one coder, the replicability and reliability of the coding could not be con-
trolled with inter-coders’ agreement metrics. Had the table been longer, it would have
made sense to feed it to software for statistical analysis and visualization, but in my
corpus, the number of photos is too small for statistics (Figure 8).
485

Meanings of tagging and captioning

Tagging is an activity that provides rich ground for observation because the uses and the
meaning of tags appear to be far from trivial. An important point that Daer, Hoffman, and
490 AQ5 Goodman (2015) have correctly observed is that much of tagging these days is not related
¶ with findability but rather has to do with rhetoric. According to these three authors, the
use of hashtags as ‘meta-commentary’ (as opposed to using them to ‘mark topics’,
organize, sort, and generate machine-readable classifications) is a distinctly rhetorical
practice (Daer et al., 2015, p. 12). The users are, thus, adapting the medium for their expres-
495 sive needs. Vernacular creativity in language use at tagging is inseparable from captioning.
In our small corpus, hashtags and their combinations become part of captions or even
 12 I. UTEKHIN

500

505
Colour online, B/W in print

510

515

Figure 7. A photo from Explore Location set. Comments in Ukrainian: [Author:] Greetings from Shishaki.
[Follower:] Oh my God! … She’s holding that man in her hand … crazy girl!!! [Author:] That’s what
archaeologists do.
520

substitute them, and often include emoji. Cf. the caption in an advertisement of an event
targeted at a very specific group, namely, the Gymnasium graduates at their graduation
day:
525
Трішки #гумору Життя воно таке – непередбачуване

Сьогодні вас буде багато, тож ми готуємо кухню і трішки розваг для вас, дорогі #випускники
Чекаємо всіх після концерту в гімназії!

(Ukrainian: ‘A bit of #humor. [emoji] Life is that way, unpredictable. [emoji]

530
You will be numerous today, so we are preparing meals and some entertainment for you, dear
#graduates. Waiting for all of you after the concert in the Gymnasium’.)

Rhetoric function comes to the forefront when not a word, but rather a whole phrase is
made a hashtag, as in the following examples:
535
Display quotations of over 40 words, or as needed.

#вГармоніїзПриродою in harmony with Nature (Ukr)

#сама_найдорожча_моя_дівчинка my dearest little girl (Ukr)
#люблюсвоихмаленьких I love my little ones (Rus)
#колесимвдвоем we two riding togheter (Rus)
540 #этижёлтыеботинки these yellow boots (Rus)
#лучшееместоназемле best place in the world (Rus)
 RUSSIAN JOURNAL OF COMMUNICATION 13

545

550
Figure 8. A fragment of the table for meta-data.

#воскресеньеденьчудесный Sunday a wonderful day (Rus)

#дождьнамнипочем we don’t care about the rain (Rus)
#дядяСашадогоняй uncle Sasha catch up with us (Rus)

555 Some of them are addressed to particular followers or a narrow group who can under-
stand private messages sent in public and who know, for instance, who is ‘uncle Sasha’.
Syntactically complex hashtags may appear in comments as well. In the following
example, the exchange involves a wordplay (Figure 9):
At the same time, the three phrases used for tags contain a segment that sounds some-
560 what similar to the word ‘fuck’, which is taken for a funny twist by both participants of the
dialogue.
Our main tag, Shishaki in its three versions, is a place name, but as a tag, it can appear
with photos that have not been taken in Shishaki, one of the possible reasons being the
fact that the photograph contains either some kind of representation of Shishaki, or some-
565 thing related with Shishaki. Thus, several photographs from an exhibition in Moscow art
gallery ‘Roza Azora’ entirely dedicated to paintings of Shishaski by Moscow artist Dudya
Sarabyanov were tagged as follows:
#moscow #downtown #rozaazora #artgallery #dudyasarabyanov #шишаки #impressionism
#russianimpressionism #art #contemporaryart #exhibition #atmosphere #atmospheric #beau-
570 tiful #smart #datcha #dacha #light #shadow #portrait #dandy #dandylad #lad #style #character
#delighted #painting

Regular meaning of place name is modified when several place names tags are listed together
in the same series. One of them might actually be a location, but the other(s) might not. It is
Moscow and Shishaki in the example above, where Moscow is used as a location tag, while
575
Colour online, B/W in print

580

585 Figure 9. Users’ dialogue in Comments. Ukrainian: yuliaka210 #I was there too gosha_goz #And me
too yuliaka210 #So I’ve seen.
 14 I. UTEKHIN

Shishaki refers to the content of the paintings at the exhibition. Several photos of the pictures
are also tagged Shishaki. Representation of representation still may receive the same tag as
the original; for instance, at some point, I saw on Instagram a photograph tagged
#шишаки (Cyrillic #shishaki) that represented myself against the background of a slide
590 with Instagram photographs of Shishaki while delivering a presentation at a conference in
Moscow; the other two tags were #research and #doublehermeneutics (Russian). This epitom-
izes the modification of the field that occurs as result of research activities.
A place name tag easily becomes a metonym, as in the case when a photo represents
schoolchildren visiting a workshop at the National Pottery Museum in Opishne. The
595 caption goes like this:
17.05.2016 #шишаки в #Музей гончарства разом з #керамленд @opishne #opishne
#Опішне

(Ukrainian: #shishaki in #Pottery Museum together with #Ceramland @opishne #opishne

600 #opishne (Ukr))

Hence, ‘Shishaki in Pottery Museum’ means that #shishaki, in this case, is about where the
schoolchildren had come from, not about the location where the photograph was taken.
In a selfie by a tourist with a backpack captioned ‘Everything is packed up. A mega-tour
begins! Let God protect me!’ the author traced his route plans across several places listed
605 in a series of tags among which Shishaki is also mentioned. The picture was taken at the
starting point of the itinerary, the city of Dnepropetrovsk.
Village or town names in their locative meaning often co-occur with ‘Ukraine’. More
than one place name is sometimes listed, as in a picture containing a screenshot of a
navigator map where the tags were as follows: #Lobkovabalka #Shishaki #Khorol
610 #Ukraine. Co-occurrence with higher level place names such as Poltava, Poltava Pro-
vince, or Ukraine, does not modify the meaning of Shishaki. Tagging Ukraine denotes
that the author is either addressing the message to potential audiences that might
be unaware that Shishaki is in Ukraine, or is wishing to emphasize the fact that this
is Ukraine and to express some emotion with respect to the content of the photograph
615 associated with the place. This might be pride, but also, rarely, condemnation, expressed
by other tags, cf.
От що зробили із лісом, у якому я раніше збирав гриби. #ліс #полтавщина #шишаки #ліс-
ництво #нелюди #україна #гриби

620 (Ukrainian: ‘That is what they’ve made with the forest where I used to collect mushrooms.
#forest #poltava province # shishaki #forestry #nonhumans #ukraine #mushrooms’)

While collecting the pictures from Instagram, I noticed that in a significant portion of
accounts, there were a number of photographs that represented easily recognizable
locations in Shishaki, but had not been tagged correspondingly. I did not add them to
625
my small corpus, nor could they be found with Instagram search by location. Many
such photos belong to students of the local Gymnasium who tag with #Shishaki not all
pictures taken here, but only the few images where mentioning the location means some-
thing more than simply indicating the place. Since the posts are addressed to the circle of
followers who easily recognize familiar surroundings, tagging the place as Shishaki, or pro-
630
viding location information would be redundant.
 RUSSIAN JOURNAL OF COMMUNICATION 15

What pragmatic meaning is possible for Shishaki to motivate tagging? Consider a boy
who was part of a team winning an amateur comedy show competition taking place on TV
channel studio in Poltava. He published a selfie with two teammates, standing before the
TV channel’s building, with the following caption and tagging:
635
У всіх ранок починається з кави, а у нас – з посмішок на ТРК ‘Лтава’

#мамаявтелике #бонжур #шишаки #квн #лигасмеха #тела #телевизор #телевидение #смех

#юмор #тольконаверх

(Everyone’s morning begins with a coffee, and ours’, with smiles at Ltava TV Channel (Ukr)
640
#mama i’m on tv #bonjour #shishaki #kvn4 #laughter league #bodies #tvset #television #laugh-
ter #humor #one way up (all tags are Russian))

Note that the city of Poltava where the event took place is not mentioned. Tagging Shishaki
here is not just showing the origin known to members of their social network; it can express
645 their being proud of having participated in the event, and, at the same time, somewhat
reminds us of how a visitor leaves a trace in the form of an inscription on a wall stating
her home place. Creative tagging is made by hand and is much more common than soft-
ware-assisted tagging which is very rare in my sample, apart from some cases where the
tags have been added by Vsco editing software.
650 Captions and hashtags in my small corpus are an attractive object for sociolinguistic analysis,
as they reflect the language situation and particularly the fact that both Ukrainian and Russian
languages are common in this area. Authors use either Russian or Ukrainian, and codeswitching
within the same posting is relatively rare, that is, the two languages do not often meet in a same
caption. Hashtag series are more prone to codeswitching. Thus, a picture representing the
655 author of the post near some tents is provided with a following caption:
Продовжую зберігати спокій ГЕСу і всього Шишацького району

Всім літа

660 #відпочинок #отдых #палатка #нашестямурах #фраер #лето #каникулы

(Ukrainian: ‘I keep on preserving the calm of the Water-power plant and of the whole Shishaki
district. Wishing [a good] Summer to everyone’ #recreation (Ukr) #recreation (Rus) # tent
(Ukr=Rus) #invasionofants (Ukr) #cool guy (slang Ukr=Rus) #Summer (Rus) #vacation (Rus))

This is also illustrative of what Daniel Miller et al. (2016) calls ‘scalable sociality’. Most
665
people who belong to the author’s social network know that there is a recreation zone
and a beach near the waterpower plant, and a few of them even possess common experi-
ence of fighting ants that were invading the tents on that occasion. We can see that the
caption and hashtags can be addressed to more or less specific audience, remaining partly
or entirely obscure to strangers.
670
English hashtags often appear together with Ukrainian or Russian ones, apparently in
order to make the post more visible. English tags do not always copy Russian or Ukrainian
tagging but rather complement it, cf.:
#summer #relax #love #life #friend #отдых #Шишаки #речка #тишина
675
(Cyrillic tags are Russian words for ‘recreation’, ‘Shishaki’, ‘river’ and ‘quiet’)
 16 I. UTEKHIN

or a following tag series with a photograph of an open air stage bearing ‘Shishaki’ written
with huge characters on the background, published by a girl who had moved elsewhere:
#Шишаки #Shyshaky #cage #home #memories #newlife #oldplace #Ukraine #клєтка

680 (#Shishaki (Ukr) – 7 English hashtags – #cage (Ukr))

‘Cage’ might mean the place represented in the photo called Cage by local young
people because it is surrounded by a fence made of metal mesh, or the cage could be
metaphoric, that is, Shishaki.
685

Conclusions: what did we learn about Shishaki?

Here, we touch upon the key question of our inquiry: what is special about Shishaki that
we come to know as a result of analyzing pictures, captions and tagging? It is reasonable to
690 suppose that feeling happy in parents’ home is something if not universally met across
Instagram locations, but very common. However, in my small corpus of data, this sweet
home message seems to occur noticeably more often than in my regular Instagram news-
feed where I have hardly ever met such a post. Does it mean that Shishaki is especially rich
in happy family life?
695 At least two plausible explanations can be offered, and not one of them is directly
based on or can be proved with the photographs. First, the authors are few, and
some of them have contributed a significant part of the pictures which convey
similar kind of message. Second, Instagram users tend to be young: mostly students
and young adults with a college degree. Young people from Shishaki get a secondary
700 education in their hometown, but usually go to college and build their career in neigh-
boring cities such as Poltava, Kharkiv, or Kyiv. They only come to Shishaki for holidays to
visit their relatives, or at least their Instagram accounts show this. The two explanations
are not mutually exclusive.
There are some more specific populations among the authors of tagged photos of
705 Shishaki:

(1) Students who come in groups for veterinary and archaeological practice or for cultural
events from Poltava;
(2) Tourists who camp by the river;
710 (3) Local schoolchildren.

The overwhelming majority of photos with Shishaki hashtags have been taken outside,
with few exceptions either posted by the café which represent parties and meals, or
showing the events that take place in the Gymnasium. Home, work, and school interiors
715 appear at the background of several Explore Location pictures. Since I operated small
data without GPS tags and cannot project the corpus onto the map, I could only recognize
some of the locations and estimate different ‘visibility’, or selective mapping of the space.
Of the many public places of the town, very few come to show in Instagram photos: Gym-
nasium and public park appear here, but not markets, stores or banks, or any other place in
720 the town center. In the few photos taken in the town, local teenagers interact with monu-
ments, attend the open-air discotheque and events that take place at the local community
 RUSSIAN JOURNAL OF COMMUNICATION 17

center. Most of the photos with recognizable location or background represent natural
landscapes, scenes at the river, the river itself, beautiful vistas from the hills, roads, and
sunsets. Hashtags by the users and my own Mood tags emphasize natural beauty,
quiet, rest, and summer, as almost a half of the tagged pictures have been taken in
725 summer. The words ‘travel’ and ‘tourism’ repeatedly appear as tags.
We could further ponder on the kinds of activity associated with certain locations and
the moods linked to the photos of these activities and taken in those locations. All of that
would not add much new information to what a person from Shishaki already knows, but
could be highly informative to a stranger. Instagram photos and their tags contribute to a
730 view of Shishaki as a place of pleasant leisure and beautiful nature, inhabited mostly by
very young people, showing their love and friendship, and young adults always on holi-
days. Needless to say, such a picture would be only partly true. On Instagram, we have
the photos, but get no access to the context of picture taking and picture sharing on Insta-
gram. We can conduct a sort of textual ethnography in order to study how people interact
735 in social networks, how they express themselves through tagging, captioning and com-
menting. However, these data will remain shallow without getting in touch with offline
social life. As a result of ethnography, the researcher brings from the field the knowledge
of the context required to explain the words and deeds of people under study. Meanwhile,
Instagram photographs are mostly devoid of the type of context the anthropologists are
740 accustomed to work with. However, Instagram photographs can be used as one of ethno-
graphic sources, and a small corpus that is possible to study by hand and eye can help to
direct further inquiry and to formulate questions that ethnography will be able to answer.
‘Small data’ and textual ethnography can assist us to interpret in more meaningful ways
the patterns made visible within the big data approach.
745

Notes
1. The Exceptional and the Everyday: 144 Hours in Kiev (http://www.the-everyday.net).
2. In August 2016, Meta-Morphology Lab St. Petersburg was held in European University at
750 St. Petersburg, directed by Lev Manovich and Damiano Cerrone. Around half a million
photos with geolocation data were used for the workshop.
3. By January 2017, the number of tagged pictures had grown considerably, even though some
productive users had deleted their accounts, and I could use, after filtering out irrelevant pic-
tures, 115 photos tagged #шишаки, 51 tagged #shyshaky, 24 tagged #shishaki, and 295 in
Explore Location. The overall figure is less than a sum, because the categories partly overlap.
755 4. KVN, acronym of Russian words for ‘Club of the Funny and Inventive People’, is ‘a Russian
humour TV show and competition where teams (usually college students) compete by
giving funny answers to questions and showing prepared sketches, that originated in the
Soviet Union’ (https://en.wikipedia.org/wiki/KVN).

760 Disclosure statement

AQ6 No potential conflict of interest was reported by the author.
¶

AQ7 References
765
¶
Banks, M., & Morphy, H. (Eds.). (1997). Rethinking visual anthropology. New Haven, CT: Yale University
Press.
 18 I. UTEKHIN

Boitsova, O. (2013). Lyubitel’skiye foto: vizual’naya kul’tura povsednevnosti /SPb.: Izdatel’stvo

AQ8 Yevropeyskogo universiteta v SPb. [Amateur photos: A visual culture of everyday life].
¶ Chalfen, R. (1987). Snapshot versions of life. Bowling Green, OH: Bowling Green State University
Popular Press.
Collier, J., & Collier, M. (1986). Visual anthropology: Photography as a research method. Albuquerque:
770 University of New Mexico Press.
Daer, A. R., Hoffman, R., & Goodman, S. (2015). Rhetorical functions of hashtag forms across social
¶
media applications. Communication Design Quarterly Review, 3(1), 12–16.
Hochman, N., & Manovich, L. (2013). Zooming into an Instagram city: Reading the local through social
media. First Monday, 18(7). Retrieved from http://firstmonday.org/ojs/index.php/fm/article/view/
4711/3698. doi:10.5210/fm.v18i7.4711
775
Hockings, P. (Ed.). (1974/2003). Principles of visual anthropology (3rd ed.). Berlin: Mouton de Gruyter.
Kontareva, A., Rudenko, N., & Ryzhakov, N. (in press). Predstavlyaya Ermitazh: poseshcheniye muzeya
¶
v epokhu sotsial’nykh media. In Ye. G. Lapina-Kratasyuk (Ed.), Interaktivnyy gorod: gorodskaya
zhizn’ v epokhu novykh media. Novoye literaturnoye obozreniye [Representing the Hermitage:
Visiting the museum in the era of social media. In: Ekaterina Lapina-Kratsyuk, editor. Interactive
AQ9 City: Urban Life in the Era of New Media. Novoye literaturnoye obozreniye].
780
¶ Kurmangaliyeva, M., & Skougarevskiy, D. (2017). Extra Jus: ‘Zhiguli’ protiv ‘Mersedesa’ [‘Zhiguli’ VS
‘Mercedes’]. In: Vedomosti № 4296 from April 6th.
Miller, D., Costa, E., Haynes, N., McDonald, T., Nicolescu, R., Sinanan, J., … Wang, X. (2016). How the
world changed social media. London: UCL Press.
Tifentale, A. (2015). Making sense of the selfie: Digital image-making and image-sharing in social
media. Scriptus Manet, 1, 47–59.
785

790

795

800

805

810