11 Sentiment Analysis of Inuit Place Names from the Kivalliq Region of Nunavut

Landscape archaeology took shape as a specific branch of archaeology during the 1960s and 1970s, at a time when so-called processual archaeology, with its characteristic support for scientific methods, held sway. As a field of study, landscape archaeology was initially couched firmly within paradigms of ecology and physical geography (Aston 1985; Hodder et al. 1995; Hodder 1987; Johnson 2012). Researchers explored how certain geographical features and the distribution of key resources influenced human interactions with the landscape in the distant past. Empirical data were sought to test ideas about why specific types of archaeological sites were located where they were. Armed with maps and aerial photographs, archaeologists used objective variables such as vegetation, elevation, geology, and hydrology to explain the locations of known archaeological sites, as well as to predict where additional sites might be found. Reconstructing and then mapping the spatial distributions of plant and animal communities, as well as the physical geography and hydrology of past landscapes, provided a powerful means of objectively exploring why people in the past chose to live where they did. With the rise of processual archaeology in the 1960s, archaeologists would expand on this idea by incorporating ecological concepts such as carrying capacity, habitat, and resource patches as a means of developing more robust, science-based interpretations of human-landscape interactions (Binford 1962, 1972; Trigger 2006, 396). In the eyes of many, this constituted a methodology for examining the relationship between landscape and people that was both rigorous and repeatable.

By the early 1980s, a few archaeologists were becoming increasingly concerned about the degree of “scientism” existing within archaeology. In response, symbolism, meaning, and human subjectivity became areas of interest in what became known as post-processual archaeology. The philosophical tradition of phenomenology provided an appealing alternative for the study of human-environment interactions. First introduced into British landscape archaeology by Christopher Tilley (1994) and Christopher Gosden (1994), phenomenological approaches represented a radical departure from those based on ecology in terms of both theory and method. The ideas of Martin Heidegger (1962, 1971) and of human geographers Denis Cosgrove and Stephen Daniels (1988) were inspirational to British post-processual archaeologists, including Julian Thomas (1999, 2008), Barbara Bender and Margot Winer (2001), and Mark Edmonds (1999, 2004).

Collectively, post-processual archaeologists critiqued objective and neutral views of space and time, in which Cartesian frameworks positioned landscapes as “objects” of study. Instead, they argued that perceptions of urban and rural landscapes were grounded in specific historical and cultural contexts and even in individual experiences. The notion that landscapes were subjectively constituted opened new avenues of research concerning the fundamental differences between “space” and “place.” In the resulting body of work, “spaces” are understood as geographical locations, or sites on the physical landscape, whereas “places” are imbued with people’s memories and lived experiences. Both Heidegger (1971) and Henri Lefebvre (2004) address the experiential dimension of space, whereby locations acquire meaning through the act of dwelling on the landscape. Through this process, often termed “place-making,” individuals invest specific locations with both personal and collective significance through their daily practices of living (see Barrett 1994; Ingold 1993, 2000; Tilley 1994; see also Gieryn 2000).

Archaeologists have investigated the distinction between space and place in both archaeological and contemporary contexts (see, for example, Bender 1998; Bender, Hamilton, and Tilley 2007; Knapp and Ashmore 1999). Because places are so closely bound up with lived experiences, whether past or relatively recent, people and communities often harbour deep-seated emotional attachments to places on the landscape. This is why landscapes are so often contested and defended, especially in post-colonial contexts. From the standpoint of research, however, the phenomenological orientation poses the problem of how best to identify, analyze, and perhaps even measure the strength of the connections people feel to the landscapes they inhabit.

Developing a systematic methodology suitable to the phenomenology of landscape proved, however, to be challenging. As Matthew Johnson notes, some archaeologists initially looked to the romantic poet William Wordsworth, who drew inspiration from the surrounding landscape simply by walking through it and immersing himself in it (Johnson 2012, 273; see also Ljunge 2013, 140). Tim Ingold’s anthropological studies of the cultural and historical reasons that people walk similarly explore the meanings derived from movement on foot through various landscapes (Ingold 2010; see also Ingold and Vergunst 2008). Others have examined the relationship between walking and storytelling among Indigenous peoples, including the Inuit (Aporta 2009), the Tłįchǫ Dene (Legat 2008), and Batek hunter-gatherers in Malaysia (Tuck-Po 2008). Yet, on a more general level, the subjective and time-bound nature of such perceptions of landscape, as well as the impossibility of reproducing any given experience of it, have led some to question the analytical rigour of such methods (Fleming 2006, 273–274). Contemporary studies have begun to utilize spatial technologies such as GPS to transform the ephemeral paths and patterns of human movement into graphic representations that are scalable, accurate, and georeferenced. Although the use of these and other digital tools, such as viewsheds and augmented reality, can be viewed as attempts to make the methodology of “phenomenological walking” more analytically robust, they can only do so much.

In another line of inquiry, anthropologists and archaeologists interested in how landscape is experienced in Indigenous cultures have focused attention on the knowledge bound up in place names and oral histories (Basso 1996; Chapin, Lamb, and Threlkeld 2005; Cruikshank 1981; Henshaw 2006; Keith 1997, 2004; Nuttall 1992; Stewart et al. 2000; Stewart, Keith, and Scottie 2004). Research into language and place names has revealed the intricate and multilayered relationships that exist between Indigenous peoples and the physical and cultural landscapes they inhabit, in both the past and the present (Bennett and Rowley 2004; Campbell 1997; Cruikshank 1990; Helleland 2006; Lyons et al. 2010; Müller-Wille and Weber Müller-Wille 1989–1991).

Our own work in the Inuit community of Arviat, on the western shore of Hudson Bay, reveals that place names are a rich source of information about how people living in the Kivalliq Region experience the coastal and inland landscapes. The people who inhabited and named these places are descendants of a group collectively known as the Caribou Inuit, a term first used by ethnographers of the Fifth Thule Expedition (1921–1924), given the group’s reliance on barren-ground caribou as a primary resource. The Caribou Inuit comprise many rather loosely affiliated groups, such as the Paatlirmiut, Nuvurugmiut, Ahairmiut, Kivihiktormiut, Qainirmiut, Hauniqtuurmiut, and Harvaqtuurmiut (Arima 1984; Birket-Smith 1929; Burch 1978, 1986). Evidence of rich and varied lifeways can be found in numerous locations in the region. In figure 11.3, we see the outlines of tent rings, caches, and hunting blinds that have been excavated out of glacial till. These stone features are situated on an island in Maguse Lake called Ikirahak, the site of a major inland caribou crossing. Figure 11.4 shows qajaq (kayak) stands along the Maguse River, where the frames of these iconic watercraft would be placed when the qajaq was not in use.

Traditional place names are highly valued within the Arviat community for many reasons. On a practical level, they provide important information that allows knowledge holders to orient themselves on the landscape, thereby aiding in navigation and travel. For example, the name Matugijjat, meaning “steep hills, an obstacle for going around,” describes a geographical feature that might make travelling through this area more challenging. Other names orient a person by referencing river flow or wind direction, such as Hannirut, which means “the island is facing north.” Organizations like Rankin Inlet Search and Rescue recognize that such knowledge facilitates people’s ability to safely navigate the land, and therefore actively advocate the teaching of place names and related information in schools.

As our conversations with Arviarmiut (that is, the people of Arviat) revealed, place names also mark the deep-seated personal and emotional connections that people have with places on the land. While some place names identify excellent harvesting and camping locations, others offer warnings about the presence of malevolent animals or supernatural beings or note significant events and associated individuals. If place names are intrinsically linked to landscape, then they represent a potentially rich data set for exploring the phenomenological dimensions of landscape. The challenge is finding a methodology with the potential to measure the strength and direction (positive or negative) of the connections that human beings have with the landscapes they inhabit.

It is evident that the nature of Inuit place names and the functions they serve express a wide range of opinions or sentiments about the land and its relationship to individuals and their cultural values. For these reasons, many Inuit place names carry an affect—an underlying emotional valence that is manifested in a person’s feelings about a place or an event. For example, a place name acquires a negative valence if the location it refers to provokes fear or disgust or is perhaps associated with physical and/or spiritual danger (see, for example, Burch 1971; Grønnow 2009; Kilabuk 2011). Similarly, a place name that provides needed information about where to collect water and food educes feelings of comfort and security. This suggests that the deeply seated emotional connections people have to the landscapes they inhabit are reflected in place names, albeit to varying degrees.

When Arviarmiut describe a particular place, one can usually determine fairly easily whether they have a largely positive or negative impression of it. However, such assessments become far more complicated when one is faced with hundreds or even thousands of place names and translations. What if computers could use algorithms to classify large numbers of place names as “positive,” “negative,” or “neutral”? Machines cannot understand natural language. However, by means of natural language processing algorithms, computers can be programmed to recognize key words within a given translation and then identify the emotional associations of a particular place. When these key words appear in a document, sentiment analysis algorithms assign a score that reflects the overall levels of positive and negative sentiment being expressed. More complex algorithms break down statements into a string of individual words and then use lexical libraries, or “bags of words,” to assign a positive and negative polarity to a particular statement. Given the complexities and organic nature of human language, computational linguists assume that most sentiment analysis algorithms are about 80 percent accurate. Nevertheless, an increasing number of businesses are using sentiment analysis of microblogging sites such as Twitter as guidelines for understanding how people feel about their products or services.

In our study, we were interested in exploring whether sentiment analysis could be used to identify places that elicited positive or negative emotions from Arviarmiut knowledge holders. We began by compiling a large database of place names from the Kivalliq Region of Nunavut, entered into an Internet-based Web application called Arctic IQ. This online electronic atlas of Inuit place names was developed in 2013 by two of the authors of this chapter, Peter Dawson and Kenneth Buck, to compile, archive, and manage a large database of place names collected during the most recent International Polar Year (2007–2008). The place names, along with their translations, were solicited in collaboration with knowledge holders living in Arviat, a small community located on the western shore of Hudson Bay. Louis Angalik, Donald Uluadluak, and Mark Kalluak worked together with the authors to enter place name information directly into the electronic map sheets contained within the Arctic IQ website and database.

Participants in the Arctic IQ project met regularly at the Nunavut Department of Education Building in Arviat to discuss place names. Working together, they effectively pooled their collective knowledge to reach a consensus on each place name as well as its location and associated meaning. The advantage of this approach is that everyone has a seat at the table. Knowledge keepers who feel they lack expertise in certain areas, for example, can still participate and contribute to the overall project. The challenge, of course, lies in assessing the reliability of the data that are collected (Brabham 2013). As with sites like Wikipedia, the basic assumption with Arctic IQ is that place names are deemed reliable when consensus on location, spelling, and meaning has been reached among the collaborators. Regardless, because errors are likely unavoidable, we consider the data contained within Arctic IQ as akin to “opinions” rather than hard facts. As opinions, they are informed by oral history and traditional knowledge, as well as by the lived experiences of their Inuit contributors. When considered in this way, it should be possible to explore the polarities of feelings and thoughts about landscape using a tool like sentiment analysis.

A natural language processing service called Lexalytics Semantria API was used to analyze the sentiments expressed by Inuit knowledge keepers relative to the 1,031 place names and their translated meanings contained in the Arctic IQ database. Each call generates a series of response fields for each place name:

Status: success/failure status indicating whether the request was processed

Language: language of the source text as detected

Type: sentiment polarity: “positive,” “negative,” or “neutral”

Score: sentiment strength (0.0 = neutral)

As an example, running place name translations through Semantria might generate negative scores of (–) 3.45724 and (–) 1.47295 for two place names, based on analysis of the words used to describe the place and its associations with events or history. In this instance, both scores are less than zero, indicating that each place name would be categorized as “negative,” with the lower score conveying a stronger negative opinion. Once the sentiment scores had been calculated for the 1,031 place names, each place name marker on the Arctic IQ map sheet was colour-coded along a sliding scale, from red (highly negative) through yellow to blue (neutral) and finally green (highly positive) (figure 11.5).

The resulting scores for Arviat place names range from highly positive (+) 0.7500000 to highly negative (–) 0.7500000. As there is no absolute way to measure the polarity of sentiment scores against a universal standard, we simply compared place name scores relative to other scores within the same data set. For our discussion, we have chosen to analyze five of the highest-scoring place names in each position on this “continuum of opinion.” The names, translations, and sentiment scores are summarized in table 11.1 and discussed in the following section. The three columns in table 11.1 provide the place name, its description, and the sentiment analysis score. A theme is also listed below the place name description. The themes created by Semantria identify topics of discussion and are determined by analyzing the context of the entity (or entities) referred to in the place name description. For example, certain words, such as “bark,” can have multiple meanings. As a noun, “bark” refers to the outer covering of a tree, while, as a verb, “bark” refers to the sound a dog makes. Semantria distinguishes between the two by determining the context in which the word is used. The result is then summarized as a particular theme, as in “dogs barking” or “tree covering.”

The colour-coding of place names by their sentiment scores allows the researcher to visualize how place name sentiments are distributed across the landscape and then search for meaningful patterns within the resulting distributions (figure 11.5). We were particularly interested in exploring whether factors like inland versus coastal positioning might produce detectable trends in sentiment polarity. By way of example, archaeologists believe that sometime during the eighteenth century coastal-dwelling Inuit groups on the western edge of Hudson Bay abandoned the relative economic security of sea mammal hunting and moved inland to hunt caribou—a resource that is far less reliable (Burch 1978, 1986, 1988; Friesen and Stewart 1994). This observation is supported by several well-documented famines that occurred in the area, following the disappearance of caribou herds (Fossett 2001, 192). If inland areas are more prone to food insecurity than coastal locations, then we might expect to see a directional trend toward place names expressing negative sentiment polarity as we move away from the coast.

Ethnohistoric accounts also indicate that tensions occurred sporadically among Inuit and First Nations groups in inland and coastal areas of southwestern Hudson Bay (but see Csonka [1994, 1995] for an opposing view). These tensions may have been heightened with the emergence of the fur trade, as well as the establishment of Fort Prince of Wales on the Churchill West Peninsula in 1717, as both Inuit and Chipewyan Dene entrepreneurs attempted to establish themselves as traders. Fossett (2001, 106) recounts one incident in which a party of Chipewyan attacked and killed several Inuit families near Knapps Bay (presently Arviat). Several place names contained in the Arctic IQ database describe skirmishes with Chipewyan at much smaller scales, all of which score negatively and appear confined to the southern portion of the Kivalliq Region. Consequently, if encounters with First Nations groups increased as Inuit moved south, then one might expect an associated trend toward named places expressing negative sentiment polarity in this area.

To explore the possibility that there might be directional trends in how opinions about places were distributed across the landscape, ArcGIS v.9.3 was used to produce an interpolation of the sentiment scores (Gillings 2012). This approach estimates values for points on a plane where no values are known on the basis of points for which values are available. In this case, interpolation offered a means to produce a simple visual representation of the extent to which the sentiment scores formed clustered or linear patterns across the landscape. Specifically, the method was used to produce an isopleth map displaying spatial variation in sentiment as a series of colour-coded zones or bands corresponding to a gradational scale of value ranges for these sentiment scores (Houlding 2000; Lu and Wong 2008). This map, in turn, provided a simple visual tool for exploring how opinions about places on the land vary based on personal experiences and local knowledge of these areas.

The isopleth map representing this interpolation surface (figure 11.6) reveals none of the directional trends hypothesized for the sentiment values within our data set. In other words, sentiment scores are not highly positive on the coast and then progressively lower and negative as one moves further inland to the west, nor do they fall as one moves south toward the zone where the traditional territories of the Arviarmiut begin to overlap with areas also used by their Chipewyan Dene neighbours. Instead, low sentiment values are distributed in localized patches, as indicated in the small, isolated areas of negative sentiment represented by green and blue patches on the isopleth map. These patches do appear to be distributed in a roughly northeast–southwest axis within the study area, a pattern that is also echoed to some extent in the loosely northeast–southwest orientation of the red and orange bands that reflect high sentiment scores.

Still, there appears to be no large spatial scale process operating here based on the results of this analysis. Rather, the landscape of the Kivalliq Region appears to be a patchwork of highly localized areas associated with positive and negative experiences. For such distinct areas to have emerged, people must have experienced these places in similar ways and then shared these experiences across generations via oral traditions and place names. Experiences of notable events (for example, murders, conflicts) would have been also passed down through the generations with the associated emotional responses. Oral traditions are the verbal messages passed along from generation to generation and that extend well beyond the living memory of the group (Mason 2000, 240; Vansina 1985, 27). Oral histories are, in contrast, the experiences one has had within one’s lifetime, and the maximum temporal scope of oral history must necessarily correspond with the age of the eldest community member (Mason 2000, 240). This is likely how the link between sentiment, landscape, and time is formed at such local levels.

When combined the interpolated ArcGIS scores, the results of the sentiment analysis provide the researcher with a visual representation of how people’s opinions of places are spatially distributed across the landscape. By identifying areas that evoke either strongly positive or strongly negative opinions, researchers are better able to engage with local knowledge holders to discover why places are perceived in a certain way. For example, do the patches of negative sentiment identified in figure 11.6 represent areas that are more prone to localized resource failures? Are they areas that contain environmental or supernatural hazards or dangers? Are they associated with historic events or persons of significance? Or are they places where supernatural beings dwell, or where important stories or myths are anchored? We hope to explore each of these possibilities through future research.

The spatial distribution of sentiment scores across the Kivalliq landscape may also provide important insights into how heritage sites are valued from an Indigenous perspective. To illustrate, the shaman’s healing stone (figure 11.7) is a site of great significance to many Arviarmiut. However, in the absence of oral histories and place names, its true meaning and value would be difficult to identify. It is therefore not surprising that heritage regulators in many provinces and territories assign value to archaeological sites based primarily on objective criteria such as age, function, and cultural affiliation. Heritage resources that fall within areas that provoke strong opinions from Indigenous knowledge keepers are likely valued for reasons that differ considerably from those emphasized by government regulators. Sentiment analysis could be used to create valuation criteria for heritage sites that take the polarity of Indigenous opinions into consideration.

Finally, we see sentiment analysis as having the potential to guide Indigenous consultation processes during proposed resource extraction and development activities. By identifying areas within a region that are likely to provoke positive or negative opinions from local stakeholders, industry representatives can work much more quickly to address community concerns.

Phenomenology in landscape archaeology represents a well-developed and sophisticated theoretical approach to understanding how people perceive the world around them. It is also an approach that requires a methodology that can yield something other than metaphorical interpretations of symbolism, meaning, experience, and identity.

While some researchers have abandoned their attempts to bridge the divide between phenomenology and technology through approaches like GIS, we propose that techniques currently being used to mine the vast amounts of data contained in social networking sites may bear fruit. Sentiment analysis, in particular, is emerging as an important area of research in computer science and computational linguistics. If place names are analyzed as “opinions” about places that are “informed” by lived experiences and cultural memories, our preliminary research shows that the different polarities of these opinions can be mined using sentiment analysis.

Our use of sentiment analysis to analyze 1,031 Inuit place names also suggests that even non-Indigenous lexical databases can identify positive and negative opinions expressed about certain places. This is not surprising, given that named places are strongly associated with deep-seated feelings derived from identity, memory, history, and mythology. Furthermore, the results of this preliminary analysis indicate that the expression of sentiment may be highly localized and structured in accordance with risk and uncertainty in the availability of key resources, historic events, and mythology, as well as geographic factors such as elevation. Our intention here is not to advocate the use of sentiment analysis as a means of “automating” the study of phenomenology in landscape archaeology. Rather, it is to explore the idea that the same techniques currently employed to “mine” the opinions of millions of Internet users might be profitably applied to toponymic research.

Clearly, there are many challenges that will have to be worked out. For example, a word that is considered “positive” in one situation may well be considered “negative” in another. The word “long” can be positive if we are describing the battery life of a smart phone, but negative if we are determining the time it takes to fully charge that smart phone. This can be further complicated when the data being analyzed have been translated from one language or cultural context into another. For example, the term “kill” has a negative association in most contexts, but a positive connotation when connected to Indigenous harvesting activities—as in a caribou kill site. This indicates that culturally sensitive lexical databases will need to be tailored to any future research.

In addition, phrases like “the book was great” and “the book was not great” indicate that even small changes in text can dramatically alter the polarity of the opinion being expressed. This could also be exacerbated when translating from a language like Inuktitut into English. Finally, most people express both positive and negative comments when reviewing a product or experience. For example, positive language could be used to describe the price of a hotel room, while the cleanliness and comfort of the room might be cast in a negative light. The same is often true for place names that describe locations where both positive and negative events have occurred.

While we fully acknowledge such caveats, the basic objective of this chapter has simply been to explore the possibility that sentiment analysis might provide new insights into how people perceive the landscapes they inhabit. Accordingly, it represents a new and novel approach to studying the phenomenology of landscape (see Eve 2012 and Hennessy et al. 2012 for examples of other new approaches). The next stage in such an analysis will necessarily involve the development of culturally sensitive lexical databases in Inuktitut. Regardless of these challenges, we believe that approaches like sentiment analysis demonstrate that technology does not have to be at odds with phenomenological approaches in landscape archaeology.

The authors would like to thank the community of Arviat for its support and hospitality over the years spanning this project. Special thanks to Donald Uluadluak, Mark Kalluak, Louis Angalik, Joe Karetak, and Luke Suluk for their generosity and for the knowledge they have shared. Thanks also to Shirley Tagalik, Natasha Lyons, and Nikki Oakden. Funding for this project was provided by the Social Sciences and Humanities Research Council of Canada, the Government of Nunavut, and the University of Calgary.

• Aniksak, Margaret Uyauperk, and Luke Suluk
• 1993 Sentry Island: An Ancient Land of Occupation. Arviat Historical Society, Arviat, Nunavut.
• Aporta, Claudio
• 2009 The Trail as Home: Inuit and Their Pan-Arctic Network of Routes. Human Ecology 3(2): 131–46.
• Arima, Eugene Y.
• 1984 Caribou Eskimo. In Handbook of North American Indians, vol. 5, Arctic, edited by David Damas, pp. 447–462. Smithsonian Institution Press, Washington, D.C.
• Aston, Mick
• 1985 Interpreting the Landscape: Landscape Archaeology in Local Studies. Batsford, London.
• Barrett, John
• 1994 Fragments from Antiquity: An Archaeology of Social Life in Britain, 2900–1200 BC. Blackwell, Oxford.
• Basso, Keith
• 1996 Wisdom Sits in Places: Landscape, Language and the Western Apache. University of New Mexico Press, Albuquerque.
• Bender, Barbara
• 1998 Stonehenge: Making Space. Berg, Oxford.
• Bender, Barbara, Sue Hamilton, and Christopher Tilley
• 2007 Stone Worlds: Narratives and Reflexivity in Landscape Archaeology. Left Coast Press, Walnut Creek, California.
• Bender, Barbara, and Margot Winer
• 2001 Contested Landscapes: Movement, Exile and Place. Berg, Oxford.
• Bennett, John, and Susan Rowley
• 2004 Uqaluriat: An Oral History of Nunavut. McGill-Queens University Press, Montréal.
• Binford, Lewis R.
• 1962 Archaeology as Anthropology. American Antiquity 28: 217–225.
• 1972 An Archaeological Perspective. Seminar Press, New York.
• Birket-Smith, Kaj
• 1929 The Caribou Eskimos: Material and Social Life and Their Cultural Position. Part 1, Descriptive Part. Report of the Fifth Thule Expedition, 1921–24, vol. 5. Gyldendalske Boghandel, Nordisk Forlag, Copenhagen.
• Brabham, Darren
• 2013 Crowdsourcing. MIT Press, Cambridge, Massachusetts.
• Burch, Ernest S., Jr.
• 1971 The Nonempirical Environment of the Arctic Alaskan Eskimos. Southwestern Journal of Anthropology 27(2): 148–165.
• 1978 Caribou Eskimo Origins: An Old Problem Reconsidered. Arctic Anthropology 15(1): 1–38.
• 1986 The Caribou Inuit. In Native Peoples: The Canadian Experience, edited by R. Bruce Morrison and C. Roderick Wilson, pp. 106–133. McClelland and Stewart, Toronto.
• 1988 Knud Rasmussen and the “Original” Inland Eskimos of Southern Keewatin. Études/Inuit/Studies 12: 81–100.
• Campbell, Lyle
• 1997 American Indian Languages: The Historical Linguistics of Native America. Oxford Studies in Anthropological Linguistics Vol. 4. Oxford University Press, New York.
• Chapin, Mac, Zachary Lamb, and Bill Threlkeld
• 2005 Mapping Indigenous Lands. Annual Review of Anthropology 34(1): 619–638.
• Cosgrove, Denis, and Stephen Daniels
• 1988 The Iconography of Landscape: Essays on the Symbolic Representation, Design, and Use of Past Environments. Cambridge University Press, Cambridge.
• Cruikshank, Julie
• 1981 Legend and Landscape: Convergence of Oral and Scientific Traditions in the Yukon Territory. Arctic Anthropology 18(2): 67–93.
• 1990 Getting the Words Right: Perspectives on Naming and Places in Athapaskan Oral History. Arctic Anthropology 27(1): 52–65.
• Csonka, Yvon
• 1994 Intermédiaires au long cours: Les relations entre Inuit du Caribou et Inuit du Cuivre au début du XX^e siècle. Études/Inuit/Studies 18(1–2): 21–48.
• 1995 Les Ahiarmiut: Àl’écart des Inuit Caribous. Editions Victor Attinger, Neuchatel, CH.
• Dawson, Peter, Richard Levy, and Natasha Lyons
• 2011 Breaking the Fourth Wall: 3D Virtual Worlds as Tools for Knowledge Repatriation in Archaeology. Journal of Social Archaeology 11(3): 387–402.
• Dorion, Henri
• 1987 Native Toponymy and Territorial Rights. Acta Borealia: A Nordic Journal of Circumpolar Societies 4(1–2): 119–126.
• Edmonds, Mark
• 1999 Ancestral Geographies of the Neolithic. Routledge, London.
• 2004 The Langdales: Landscape and Prehistory in a Lakeland Valley. Tempus, Stroud, UK.
• Eve, Stuart
• 2012 Augmenting Phenomenology: Using Augmented Reality to Aid Archaeological Phenomenology in the Landscape. Journal of Archaeological Method and Theory 19(4): 582–600.
• Feldman, Ronen
• 2013 Techniques and Applications for Sentiment Analysis. Communications of the ACM 56(4): 82–89.
• Fleming, Andrew
• 2006 Post-processual Landscape Archaeology: A Critique. Cambridge Archaeological Journal 16(3): 267–280.
• Fossett, Renée
• 2001 In Order to Live Untroubled: Inuit of the Central Arctic, 1550 to 1940. University of Manitoba Press, Winnipeg.
• Friesen, T. Max, and Andrew Stewart
• 1994 Protohistoric Settlement Patterns in the Interior District of Keewatin: Implications for Caribou Inuit Social Organization. In Threads of Prehistory: Papers in Honour of William E. Taylor, Jr., edited by David Morrison and Jean-Luc Pilon, pp. 341–360. Canadian Museum of Civilization Paper 149, Ottawa.
• Gieryn, Thomas F.
• 2000 A Space for Place in Sociology. Annual Review of Sociology 26: 463–496.
• Gillings, Mark
• 2012 Landscape Phenomenology, GIS and the Role of Affordance. Journal of Archaeological Method and Theory 19(4): 601–611.
• Gosden, Christopher
• 1994 Social Being and Time. Wiley-Blackwell, New York.
• Grønnow, Bjarne
• 2009 Blessings and Horrors of the Interior: Ethnohistorical Studies of Inuit Perceptions Concerning the Inland Region of West Greenland. Arctic Anthropology 46(1–2): 191–201.
• Heidegger, Martin
• 1962 Being and Time. SCM Press, London.
• 1971 Building, Dwelling, Thinking. In Martin Heidegger, Poetry, Language, Thought, pp. 141–160. Translated by Albert Hofstadter. Harper & Row, New York.
• Helleland, Botolv
• 2006 The Social and Cultural Values of Geographical Names. In United Nations Group of Experts on Geographical Names, Manual for the Standardization of Geographical Names, pp. 121–128. United Nations, New York.
• Hennessy, Kate, Ryan Wallace, Nicholas Jakobsen, and Charles Arnold
• 2012 Virtual Repatriation and the Application Programming Interface: From the Smithsonian Institution’s MacFarlane Collection to “Inuvialuit Living History.” In Reconsidering Barnett Newman, edited by Melissa Ho, pp. 12–21. Philadelphia Museum of Art, Philadelphia.
• Henshaw, Anne
• 2006 Pausing along the Journey: Learning Landscapes, Environmental Change, and Toponymy amongst the Sikusilarmiut. Arctic Anthropology 43(1): 52–66.
• Hodder, Ian
• 1987 The Archaeology of Conceptual Meaning. Cambridge University Press, Cambridge.
• Hodder, Ian, Michael Shanks, Alexandra Alexandri, Victor Buchli, John Carman, Jonathan Last, and Gavin Lucas
• 1995 Interpreting Archaeology: Finding Meaning in the Past. Routledge, London.
• Houlding, Simon
• 2000 Practical Geostatistics: Modeling and Spatial Analysis. Springer, New York.
• Ingold, Tim
• 1993 The Temporality of Landscape. World Archaeology 25: 152–174.
• 2000 The Perception of the Environment: Essays on Livelihood, Dwelling and Skill. Routledge, London.
• 2010 Footprints Through the Weather-World: Walking, Breathing, Knowing. Journal of the Royal Anthropological Institute 16(s1): S121–S139.
• Ingold, Tim, and Jo Lee Vergunst
• 2008 Ways of Walking: Ethnography and Practice on Foot. Ashgate, Aldershot, UK.
• Johnson, Matthew H.
• 2012 Phenomenological Approaches in Landscape Archaeology. Annual Review of Anthropology 41(1): 269–284.
• Keith, Darren
• 1997 Arvia’juaq National Historic Site: Conservation and Presentation Report. Arviat Historical Society, Arviat, Nunavut, and Parks Canada.
• 2004 Caribou, River and Ocean: Harvaqtuurmiut Landscape Organization and Orientation. Études/Inuit/Studies 28(2): 39–56.
• Kilabuk, Elisha
• 2011 The Qalupalik. Inhabit Media, Iqaluit.
• Knapp, A. Bernard, and Wendy Ashmore
• 1999 Archaeological Landscapes: Constructed, Conceptualized, Ideational. In Archaeologies of Landscape: Contemporary Perspectives, edited by Wendy Ashmore and A. Bernard Knapp, pp. 1–30. Blackwell, Oxford.
• Kumar, Akshi, and Teeja Mary Sebastian
• 2012a Sentiment Analysis on Twitter. International Journal of Computer Science Issues 9(4): 372–379.
• 2012b Sentiment Analysis: A Perspective on Its Past, Present and Future. International Journal of Intelligent Systems and Applications 4(10): 1–14.
• Lefebvre, Henri
• 2004 Rythmanalysis: Space, Time, and Everyday Life. Continuum, London.
• Legat, Allice
• 2008 Walking Stories; Leaving Footprints. In Ways of Walking: Ethnography and Practice on Foot, edited by Tim Ingold and Jo Lee Vergunst, pp. 35–50. Ashgate, Aldershot, UK.
• Levy, Richard, and Peter Dawson
• 2014 Interactive Worlds as Educational Tools for Understanding Arctic Life. In Pastplay: Playing with Technology and History, edited by Kevin Kee, pp. 66–86. University of Michigan Press, Ann Arbor.
• Lyons, Natasha, Peter Dawson, Matthew Walls, Donald Ulluadluak, Louis Angalik, Mark Kalluak, Philip Kigusiutuak, Lukle Kiniski, Joe Keretak, and Luke Suluk
• 2010 Person, Place, Memory, Thing: How Inuit Elders Are Informing Archaeological Practice in the Canadian North. Canadian Journal of Archaeology 34: 1–31.
• Lu, George Y., and David W. Wong
• 2008 An Adaptive Inverse-Distance Weighting Spatial Interpolation Technique. Computers and Geosciences 34: 1044–1055.
• Ljunge, Magnus
• 2013 Beyond “the Phenomenological Walk”: Perspectives on the Experience of Images. Norwegian Archaeological Review 46(2): 139–158.
• Mason, Ronald J.
• 2000 Archaeology and Native North American Oral Traditions. American Antiquity 65(2): 239–266.
• Mohammad, Saif, Svetlana Kiritchencko, and Xiaodan Zhu
• 2013 NRC-Canada: Building the State-of-the-Art in Sentiment Analysis of Tweets. In Proceedings of the Seventh International Workshop on Semantic Evaluation Exercises (SemEval-2013). Atlanta, Georgia.
• Müller-Wille, Ludger, and Linna Weber Müller-Wille
• 1989–1991 Keewatin NUNA-TOP Survey of Inuit Geographical Names. Archival and Electronic Data and Maps. Indigenous Names Surveys. Department of Geography, McGill University, Montréal.
• Nuttall, Mark
• 1992 Arctic Homeland: Kinship, Community, and Development in Northwest Greenland. University of Toronto Press, Toronto.
• Stewart, Andrew, T. Max Friesen, Darren Keith, and Lyle Henderson
• 2000 Archaeology and Oral History of Inuit Land Use on the Kazan River, Nunavut: A Feature-Based Approach. Arctic 53(3): 260–278.
• Stewart, Andrew, Darren Keith, and Joan Scottie
• 2004 Caribou Crossings and Cultural Meanings: Placing Traditional Knowledge and Archaeology in Context in an Inuit Landscape. Journal of Archaeological Method and Theory 11(2): 183–211.
• Thomas, Julian
• 1999 Culture, Time, and Identity: An Interpretive Archaeology. Routledge, London.
• 2008 Archaeology, Landscape and Dwelling. In Handbook of Landscape Archaeology, edited by B. David and J. Thomas, pp. 300–306. Left Coast Press, Walnut Creek, California.
• Thornton, Thomas F.
• 1997 Anthropological Studies of Native American Place Naming. American Indian Quarterly 21(2): 209–228.
• Tilley, Christopher
• 1994 A Phenomenology of Landscape: Places, Paths and Monuments. Routledge, London.
• Trigger, Bruce
• 2006 A History of Archaeological Thought. Cambridge University Press, Cambridge.
• Tuck-Po, Lye
• 2008 Before a Step Too Far: Walking with Batek Hunter-Gatherers in the Forests of Pahang, Malaysia. In Ways of Walking: Ethnography and Practice on Foot, edited by Tim Ingold and Jo Lee Vergunst, pp. 21–34. Ashgate, Aldershot, UK.
• Utok, Tony, Luke Suluk, and Darren Keith
• 1994 Maguse River Place Names Project. Field Report, Parks Canada, Department of Canadian Heritage, Gatineau, Québec.
• Vansina, Jan
• 1985 Oral Tradition as History. University of Wisconsin Press, Madison.
• Whitridge, Peter
• 2004 Landscapes, Houses, Bodies, Things: “Place” and the Archaeology of Inuit Imaginaries. Journal of Archaeological Method and Theory 11(2): 213–250.