Translating multimodal texts in space: A case study of St Mungo Museum of Religious Life and Art

Multimodality has received considerable critical attention in Translation Studies over the last decades. However, how translations interact with or within three-dimensional material space is still under researched. This study proposes to use the study of geosemiotics (Scollon & Scollon 2003) as the theoretical framework to explore this new territory. A case study is carried out with St Mungo Museum of Religious Life and Art in Glasgow. The multimodal analysis divides museum space into four ranks: the museum surroundings, the museum building, the museum exhibition, and the museum objects. The findings reveal that the translated exhibition texts interact with the four ranks of spaces to consistently minimize the narratives of Christian heritage in Glasgow and manifest a multireligious and multi-ethnic Scottish idenity. This study demonstrates how (non)provision of translations can potentially have an impact on the construction of in-place meaning in the multimodal museum space.


Introduction
Multimodality can be defined as "the use of several semiotic modes in the design of a semiotic product or event" (Kress & Van Leeuwen, 2001, p. 20). In Translation Studies, the concept of multimodal texts seems to have been mostly associated with audiovisual or multimedia texts, i.e. "the processing and presentation of information in two or more media simultaneously" (Cattrysse, 2001, p. 1). Typical examples of multimedia texts include films and television programs. A significant number of studies have been devoted to these types of texts, and many of them address the problems of technical constraints on the translation of verbal texts imposed by other modes such as images and soundtrack (see Gambier & Gottlieb, 2001;Orero 2004). Other multimodal texts have also gradually received some attention from translation scholars. Some examples are opera (Mateo, 2007), picture books (Oittinen, 2001), comics (Zanettin, 2008), and video games (O'Hagan & Mangiron, 2013).
These studies on multimodal translation have challenged Translation Studies in a number of ways (see Gambier & Gottlieb, 2001;O' Sullivan, 2013). However, compared with the wider issues that have been explored in multimodal discourse or visual analysis, it seems that so far the focus on multimodality within Translation Studies is still limited. To date, little is known about how translations interact with or within three-dimensional material space. This study attempts to explore this new territory by examining written translations in a museum site, as museums have often been chosen for collecting data for the exploration of threedimensional space (e.g. Pang, 2004;Ravelli & McMurtrie, 2015).
Museum translation is a relatively under-researched area in both museum and translation studies (see Liao 2018). Within limited existing studies, only a few have begun to explore museum communication as a multi-semiotic event (e.g. Neather, 2008Neather, , 2012Jiménez Hurtado & Soler Gallego, 2015;Sell, 2015). Physical features in museum space, such as architecture, floor plans and circulation paths, remain unaddressed by translation scholars. In this paper, a case study is carried out to examine how translation interacts with museum space and museum visitors in St Mungo Museum of Religious Life and Art in Glasgow (hence St Mungo Museum).
Specifically, this paper aims to answer the following research questions:


How do verbal texts form part of the space?  How do verbal texts guide visitors to perceive the space, and does this influence their perception of the exhibitions in the museum?  Following the first two questions, are there differences between the source and the translated texts in relation to the museum space?

Theoretical framework: geosemiotics
For this research on translated texts, we need to make a link between space and verbal texts. For this purpose, we find the theory of geosemiotics developed by Scollon & Scollon (2003) particularly relevant. Geosemiotics is defined as "the study of the social meaning of the material placement of signs in the world" (Scollon & Scollon, 2003, p. 110). This theory brings well established studies which have previously not been integrated together to form a single and coherent framework for the analysis of in-place meaning, such as discourse analysis, social psychology, communication studies, and cultural geography. Specifically, four components constitute the system of geosemiotics: social actor, interaction order, visual semiotics and place semiotics. Social actor is "the habitus of individual beings" (Scollon & Scollon, 2003, p. 166). Each individual located at a certain place reading a particular text in a certain environment has their history -e.g. the languages they speak, the place they come from -and these influence how they interpret the texts on a particular occasion. In the museum settings, different people standing in front of a displayed object reading Translating multimodal texts in space the same label can create different narratives of the exhibition, because of their personal background. It can even be argued that the habitus of these visitors shapes their museum experience before they even visit the museum.
Interaction order is a concept taken from Goffman (1983). It refers to the conventional sets of social relationships, which we may observe or challenge, and this is closely related to how we see self and other in this social relationship (Scollon & Scollon, 2003, p. 16). Taking the site of St Mungo Museum to illustrate, the curator recalled how some religious believers perceived the museum as a religious rather than a secular place, and contested that religious conventions needed to be followed in terms of how to place a religious object. Disagreement with how the museum interprets certain religious practices even led to visitors damaging the exhibited objects (Carnegie, 2009, p. 163).
Visual semiotics in this framework is based on van Leeuwen (1996, 2001), and is broadly defined in this framework as "any and all of the ways in which meaning is structured within our visual fields" (Scollon & Scollon, 2003, p. 11). A range of tools from Leeuwen (1996, 2001) are employed in this study to examine the visual composition, such as modality (colour saturation, colour modulation, brightness, etc.) and composition (centred, polarized, etc.). However, an important extension of visual analysis in this theory is the focus on how visual images are placed in the real world, and how social actors index these images and thus construct their ongoing social performance. In other words, this theory is interested in "how images mean what they mean because of where we see them" (Scollon & Scollon, 2003, p. 84).
Finally, place semiotics is a term coined by Scollon and Scollon (2003), which refers to "the huge aggregation of semiotic systems which are not located in the persons of social actors or in the framed artifacts of visual semiotics" (Scollon & Scollon 2003, p.8). Examples of place semiotics include architecture, urban planning, and landscape analysis. This is the most crucial component of the four, because "the central thesis of geosemiotics is that exactly where on earth an action takes place is an important part of its meaning" (Scollon & Scollon 2003, p.18, original emphasis).
Overall, the framework of geosemiotics can be summarized as following: the in-place meaning is created through a particular human being, with their discursive background, behaving in a social relationship regarded by them as relevant and appropriate, influenced by any visual semiotics within their visual fields, from the physical place they are situated. This framework allows us to examine whether the differences of the exhibition texts in the source text (ST) and in the target text (TT) provided by the St Mungo Museum, belonging to the component of visual semiotics in the system of geosemiotics, may have a potential impact on pointing the two broadly different groups of social actors -museum visitors relying on the ST and the TT -to construct different in-place meanings.
Despite being insightful to the present study, there are some limitations in geosemiotics. First, the theory is mainly illustrated with multimodal texts containing relatively few words, such as shop signs and roadway signs. As the authors explained, "because so little language is involved that we can see fairly clearly how the different elements of the indexicality -indexable linkage are working" (Scollon and Scollon 2003, p.14). For example, the authors gave a detailed example of how the twoword traffic sign "ONE WAY" interacts with the visual semiotics of an arrow vector, and the place where it is placed, to create the meaning of "driving only in one direction". This book gives no examples of analyzing longer verbal texts or a large number of coherently linked texts, such as labels in museums. In longer texts, it is likely that not all words in texts are indexical, or the texts may index a more complex structure of places. Second, as these examples are mostly taken from open and public spaces, in contrast with the purposefully-designed spaces such as museums, the theory does not provide a systematic tool for the analysis of place semiotics. In Scollon and Scollon (2003), the analyses of place semiotics are mainly ad-hoc descriptions of contextual background relating specifically to the examples given (e.g. reading conventions in different Chinese communities). Thirdly, despite recognizing the role of social actors and their individual background in geosemiotics, no methodology is offered to examine this component, apart from some bypassing comments, for example: "to test these ideas we and our colleagues conducted focus groups and interviews in which participants were asked to discuss a variety of photographs of signs" (Scollon and Scollon 2003, p.134). These are the issues that we need to solve in this study, and will be further explained in the section of data collection and methodology.
Finally, one obvious limitation in this theory is that it does not address the role of translated texts in space directly, notwithstanding some comments on public signs in multilingual communities. We argue that how verbal texts guide visitors to perceive the museum space is particularly relevant to a study of translation, because users of translations are often tourists from other countries or members of minority community in a multilingual society, so they may not be as familiar with the museum and its related culture as the readers of the source text. In the discussion of "exhibiting other cultures", Lidchi (1991, p. 166) argues that it is more difficult for visitors to establish a common ground with objects from distant and unfamiliar cultures, so the accompanying texts play a crucial role in "navigat[ing] the reader on a directed route through potentially complex and unfamiliar terrain." Therefore, visitors who are unfamiliar with the source culture of the museums or the exhibited objects may be more dependent on interpretations provided by museums in a range of forms such as object labels, audio guides, and interactive Translating multimodal texts in space touch screens 1 . It is therefore interesting to examine whether and if so, how the source texts and the translated texts in museums point visitors to construct different in-place meanings.

Background of the museum
St Mungo Museum is operated by the Glasgow City Council. The aim of the museum is to "promote mutual understanding and respect among people of all faiths and none", as explicitly stated in the entrance foyer. O'Neil (2006) and Carnegie (2009) provide a detailed account of their involvement in curating this museum. Both point out that this museum was designed to reflect the multi-ethnic Glasgow identity: "it aims to reflect the much more complex identity of Glasgow and Scotland as a multi-faith rather than a mono-cultural society" (O'Neil, 2006, p. 40). From the curators' accounts, it can be observed that both are cautious in defining this space. O'Neil (2006, p. 42) defines this space as "a safe arena which is secular but borders the religious or spiritual domain." Carnegie (2009, p. 160) explains, "despite being adjacent to Glasgow Cathedral, […] it is a secular space which attempts to make sense of the central role and importance of religion in people's lives." Both curators clearly emphasized the word "secular", and tried to present religion in a more neutral or "museumized" manner.
Despite the aim of promoting understanding of and respect for the multi-faith community, both curators highlight the fact that this museum is a former cathedral building and located within a Catholic Square in Glasgow, and therefore the museum inevitably sends a strong Christian message. The clearest message of Christian tradition comes from the name "St Mungo", Glasgow's patron saint who brought the Christian faith to Scotland. Carnegie (2009, p. 161) admits there were concerns of whether this choice of name might discourage some potential visitors from the local community, but the decision was made by the ruling council at that time. Carnegie (2009, p. 168) further points out that despite the aim of displaying world religions, this exhibition is interpreted "from a Western perspective, to raise themes of global importance" (emphasis added). O'Neil (2006, p. 41) also makes it clear that the museum "does not pretend to be objective and value-neutral." These conflicting discourses between a traditional Christian Glasgow identity and a multi-faith Glasgow identity promoted by the museum will form the basis of the following analysis of the museum space.

Min-Hsiu Liao
For the purpose of this research, three sets of data have been collected. First, a set of field notes have been compiled based on three field trips to the museum at different stages of the research process in March and August 2017, and March 2018. During the three trips, the external and internal spatial features of the museum that have caught the author's attention were noted down and photographed. Leaflets and other useful documents from the museum were also collected.
Second, all multilingual "exhibition texts" 2 in the three permanent gallery exhibitions were photographed. For the convenience of references, in the following all exhibition texts will be referred to as "labels". English is the source language in this museum, either presented above or in the left of the translated texts. The translated languages are organized from top to bottom in an alphabetical order: Arabic, Gaelic, Mandarin Chinese and Urdu (see Figure 1). Overall, 46 multilingual labels were collected. For the analysis of geosemiotics, how all translated labels were placed in the museum space were observed. The detailed linguistic analysis will only include the English and the translated Chinese texts, these being the author's languages of specialism.

Translating multimodal texts in space
Third, in order to better understand the role of social actors, visitors' comments from three online travel reviews were collected, including a total of 266 comments from TripAdvisor; 54 reviews from Google Review; and 5 reviews from a Chinese travel forum Qiongyou, all collected on 15th May 2017. Visitors' comments are commonly used as data in audience research in museum studies (MacDonald, 2005), and recently comments on travel review websites such as TripAdvisor have been used in a number of museum studies (e.g. Owens, 2012;Carter, 2016).
It is perhaps useful to clarify here that this study does not intend to measure the effect of verbal or visual signs on visitors, as it is not possible to specifically identify the language profile of the commenters. An in-depth analysis of visitors' comments or an observation of visitors' behaviour is beyond the scope of this study. Visitor comments in this study are only intended to be used as a further set of data to triangulate our findings from the multimodal analysis of in-place meaning.

An analytical model of museum space
Since the theory of geosemiotics did not provide a systematic tool for the analysis of place semiotics, in order to analyse museum space in a systematic manner, we consult the studies of museum space (Hiller & Tzortzi, 2011;Ravelli & McMurtrie, 2015), and a systemic functional framework for museum exhibitions developed in Pang (2004, p. 58-59). Hiller & Tzortzi (2011, p. 283) points out that space is an intrinsic aspect of human activity, not just background; and meaning is constructed not only through properties of but also relations between spaces. The museum space not only refers to its properties, e.g. the architecture style, the material of the building, but also relations between spaces, e.g. how one gallery leads to another, and how visitors move around in the museum. To be more concrete, Pang (2004) divides exhibition space into four categories for analysis: a museum (as a site), a gallery (as an exhibition space), an area (as a particular exhibition room), and an item (as a particular displayed unit).
Pang's categorization is useful, but its emphasis is more on the internal space of a museum, rather than external space. In the case of St Mungo Museum, as observed in the curators' discussions above, the surroundings of this museum play a crucial role in determining how it may be perceived by the visitors. Thus, we decided to divide Pang's category of "museum (as a site)" into the sub-categories "museum surroundings" and "museum building". Furthermore, since St Mungo Museum is a relatively small museum with only three galleries, Pang's categories of "gallery" and "area" are combined into one category "museum exhibition." After the four categories were decided, the notes on the spatial features of the museum taken during the field trips were examined, and Min-Hsiu Liao these features were attributed to an appropriate category by the researcher, according to the description of each rank of space as explained in Table 1.

Museum Building
This rank concerns the museum building itself, including its external architectural appeal, the properties of the building, or simply what the building looks like from outside.

Museum Exhibition
This rank relates to the internal layout of the museum, including the floor plans, the exhibition rooms, the circulation path, etc. These spatial features affect how visitors move around in the museum from one location to another.

Museum Objects
This rank addresses how an object is displayed in relation to other semiotic modes, and how they interact with the above three ranks of space.

An overview of the museum space
This section begins with an overall analysis of the four ranks of space in St Mungo Museum, with a particular view of how the contested religious discourses are embedded in space. In terms of museum surroundings, the museum is located in Glasgow's Cathedral Square, and the nearby buildings include Glasgow Cathedral, Glasgow Necropolis, and Glasgow Evangelical Church. The museum building was established in 1989, originally owned by Glasgow Cathedral. It was converted into a museum of religions in 1993. The architecture emulates Scottish baronial style, reflecting the architecture of the Bishop's Castle, a medieval castle in Glasgow.
The internal structure of the museum is divided into three floors and four exhibition galleries. The ground floor contains an information desk, a shop, and a cafe. On the first floor, the first gallery is the Gallery of Religious Art (see Figure 2 3 ). It is a bright and open space, with a display of a selection of religious and folklore objects. A smaller room Translating multimodal texts in space adjacent to this gallery is the Gallery of Religious Life. This room is dimly lit, and the exhibition cases are densely structured along a Ushaped pathway (see Figure 3). The room is structured in such a way that visitors are guided to follow a fixed pathway. The second floor houses a temporary exhibition room. The Scottish Gallery on the third floor is an open and bright space, with some hands-on activities for young visitors. In terms of the arrangement of displayed objects, in the Scottish Gallery and the Gallery of Religious Life, the objects are grouped thematically according to religions. A thematic label usually accompanies an exhibition case. In the Gallery of Religious Art, aretfacts are displayed one by one and each with an object label.
The above analysis shows that the museum surroundings and the museum building are clearly characterized by a Christian spatial discourse (e.g. being located in Glasgow's Cathedral Square). Many of these spatial features can be seen by visitors before they enter the museum, and probably prompt visitors to form an expectation of what the museum is about. The internal design sends a stronger message on the multi-faith value in Glasgow, with a mix of religious and folklore objects from the world displayed. How different religions are conceived and practised by the followers inhabiting Glasgow is explained. The ways that objects are organized religion by religion, and the religions are presented in an alphabetical order, seem "neutral" enough and reflect the aim of respecting all religions. Nevertheless, a close examination of the exhibition rooms reveals some interesting spatial features.
In the Gallery of Religious Life (Figure 3), for example, six religions displayed in the thematic cases in the middle of the room are surrounded by a cycle of life, which although can be taken as universal, arguably embodies decisions made by the institution, from the curatorial perspective -for example, the concepts of "afterlife" and "go-betweens" are not shared by all religions, and the decision to have an exhibition case on "persecution, war and peace" strongly reflects an institutional ideology, rather than a universal or neutral religious discourse.
The layout of the Gallery of Religious Art also consolidates the view that this is a religious museum from the Western perspective. As can be seen in Figure 2 The internal space of the museum exhibition is also linked to the surroundings, and this is perhaps most explicitly designed in the Scottish Gallery, where a window invites visitors to view the surroundings, and a label explains the six important buildings nearby. Figure 4 shows the view through the window. Overall, the analysis of the four ranks of museum space seems to reveal that the spaces of the four ranks all contribute to both the discourse of a Christian and a multi-faith Glasgow. In the space, the textual guides also help make some spatial features more explicit, for example, by pointing visitors to the cathedral surroundings.
We now attempt to triangulate this multimodal analysis with some recurrent themes that emerge from the visitors' comments. One of the themes is reference to the surrounding buildings of the museum, particularly the Catholic Cathedral. For example: "[the museum is] situated beside the Cathedral, close to the Necropolis and across the road from the Provand's Lordship." Some comments further indicate that the museum's surroundings is the reason for visiting: "very close to Cathedral and the Necropolis so well worth calling in here if you are visiting." The reference to architecture style also appeared in visitors' comments. It is interesting to note that several comments refer to the building as the church, and one comment refers to the impact of the architeucture on the museum experience: "being in here is a bit like entering a church you are in awe of the place". One comment further illustrates how the external features of the museum form the expectation of the visit: "the appearance of the building does not feel like dealing with various religions." In terms of comments on the collection, there are mixed opinions. Many comments contain keywords that reflect the aim of the museum, such as "muliticuture", "world religions", "various religions", "all major religions." On the other hand, a number of comments perceive the collection as mainly Christian -for example: "The collection is mostly but not entirely Christian." As stated above, it is not possible for us to measure the effect of the place semiotics on visitors' experiences. However, the brief overview of the comments support some of our observation and arguments based on geosemiotics. First, the spatial features play a clear role in visitors' experience, even before they start their visiting experience. Second, the conflicting discourses between a Scotland of Christianity or multi-cultural religious heritage also appeared in visitors' comments.
The next section further investigates the role of translations in the space of St Mungo Museum.

The translated spatial text
An initial observation of the multilingual labels in the museum shows that the source texts are not entirely translated. The Gallery of Religious Life has wall labels accompanying each exhibition case, and all are translated, with some shifts at micro-level such as lexical choices. There are also some labels accompanying individual objects inside the exhibition cases, but these are only available in English. The Gallery of Religious of Art has object labels accompanying individual artefacts, and the translated labels are shorter than the source text, of varying degrees. The Scottish Gallery has thematic labels explaining each religion, but only some are translated. Below, how translations interact with the four ranks of museum spaces will be examined.

Museum surroundings and buildings
As discussed above, one prominent spatial feature of the museum surroundings is being adjacent to Glasgow Cathedral and other buildings with Christian connections, and this is also noticed in the visitors' comments. However, how the museum is related to its surroundings, as indicated in the verbal texts and objects in the museum, seems to have

Translating multimodal texts in space
shifted in the translated text, because of the selective provisions of translations in the galleries.
This selection is particularly clear in the Gallery of Religious Art: all the objects in the outer circle on the three walls only have the title of the label translated, and all these objects happen to be related to Christianity and Glasgow's history (see Figure 2). Examples include: "The Bible: Roses", "Old Testament figures", "Angels", and "Friends and followers of Christ". These are all stained glass panels taken from churches in Glasgow, and show distinctive features in Glasgow. These displayed artefacts interact with the museum architecture to emphasize the religious tradition and heritage in Glasgow. On the other hand, all the objects in the inner circle, including religious and folklore artefacts from China, Australia or Africa, have longer translations -still reduced, but more than just titles.
Another spatial feature that highlights the museum surroundings, as indicated above, is a window on the top floor of the museum. The label next to the window explaining the surrounding buildings can be considered as an invitation to encourage visitors to link the internal and the external spatial features. However, this label is not translated, and therefore these visitors are not introduced to the Christian buildings in the museum's surroundings.

Museum exhibitions
The selective translation in the Gallery of Religious Art as discussed above does not only affect visitors' perception of the museum's surroundings, but also how the museum texts can potentially guide visitors to move around in the exhibition room. The layout of this gallery is relatively open-plan, as demonstrated in Figure 2 above, so we can probably argue that the placement of museum texts encourages visitors to move around in the open space to choose objects that they want to look at more closely or to learn more about. However, the content of the object labels on the three walls of the gallery are not translated. We argue such design of place semiotics through non-provision of translations may prompt the visitors to produce a different in-place meaning, i.e. pointing visitors to skip the objects in the outer circle, or to be less engaged with them. We may imagine that in terms of the physical movement in this space, the visitors who are completely relying on the verbal texts for the knowledge of the objects will either be diverted into a more fixed circulation path (only the inner circle), or the flow of movement in the outer circle may speed up simply because there is less to read, and they are not encouraged to look closer at the objects. The visitors thus have the potential to be diverted to world religious objects only, and their contact with Christian objects may be reduced. This illustrates how placement of (non)translations in space can potentially influence what visual semiotics fall within the visual fields of social actors, and thus influence their construction of in-place meaning.

Museum objects
Museum labels guide visitors to look more closely at objects, and one effective way to engage visitors is directing them to look at specific points on the object. Below we will use an object in the Gallery of Religious Art as an example to explain how the object is linked to the other ranks of museum space and the aim of the museum ( Figure 5).

Translating multimodal texts in space
A label is placed at the left-bottom side of this window. The English text contains 298 words, including the title "OLD TESTAMENT FIGURES"; three subsections explain the three glass panels from left to right: a) The young Samuel; b) Elijah and the Fiery Chariot; and c) Samuel as an old man, and a concluding paragraph highlights how this stained panel exemplifies a Glasgow technique called "the cameo technique". In terms of the object-text interaction, the English text directs visitors through the space on the displayed unit -from left to right, following the order of the subsections, and pointing visitors to specific areas on the window, for example -"the inscription at the top of the window". At the museum level, the label highlights the Glasgow heritage and historical context, with phrases such as "the window was made for Old Partick Parish Church in Glasgow by William Meikle & Sons in 1905." These references point visitors to the wider space the museum inhabits, i.e. Glasgow city.
The languages are organized in two columns. A long English text on the left, and the four translations of only the object title on the right, listed from top to bottom in alphabetical order: Arabic, Gaelic, Mandarin Chinese and Urdu (see Figure 6). The visitors relying on the translated label can infer from the words "Old Testament" that this displayed unit is related to Christianity, but the spatial link to the details in the stained window and to the wider context of Glasgow heritage is lost. They are also not oriented through specific features on the window, i.e. the space of the object. This is why their engagement with the objects may be lower, and they may move more quickly from one object to another, or probably skip the objects altogether.
Understandably, providing translations for all labels may not be possible for museums due to various constraints, e.g. cost and space. However, deciding what to translate is always motivated. Ravelli (2006, p. 121) comments that 'one exhibition can facilitate particular forms of visitor interaction, can prioritize some meanings in the exhibition rather than others, and can construct a picture of what the subject matter "is"'. Based on our analysis, the subject matter in the translated spatial text seems to be displaying 'world religions' or the multi-faith Glasgow identity, rather than the Glasgow heritage and Christian tradition.

Discussion: translated labels in gemosemiotics
In the analysis above, we have identified possibilities of how a consistent pattern of selective translations may have an impact on the spatial text. Overall a consistent pattern in the four ranks of space to move from Christian tradition to a museum of world religions has been identified, and how museum texts and spaces at different ranks interact with each other has been demonstrated. To explain the role of translation in the framework of geosemiotics, we can perhaps view translated labels as a physical object under the cateogry discussed in Scollon and Scollon (2003) as visual semiotics. The translated labels as physical objects attract visitors who are unfamiliar with other visual semiotics in the museum space, including the displayed artefacts from distant cultures and labels in an unfamiliar language. The translated labels attract visitors to the associated visual objects, such as an exhibition case or a displayed item. The translated labels also function to point visitors to other features of place semiotics such as the surroundings of the museum or the history of the city, which may otherwise be ignored by visitors.
On the other hand, the non-provision of translated labels may fail to draw visitors' attention to the associated visual semiotics or place semiotics that the source texts point to, and therefore minimize the semiotic resources that visitors can refer to for their construction of inplace meaning. If the non-provision is carefully designed by the museum to manifest a particular narrative in the museum, then the visitors who rely on the translated labels are likely to be guided to form a different interpretation of the in-place meaning from that of the source text readers, as illustrated in this case study.

Conclusion
This paper applies the theory of geosemiotics to multimodal translation studies. We have demonstrated that geosemiotics extends the research focus from visual semiotics to place semiotics, by looking at where visual Translating multimodal texts in space semiotics (including artifacts, the ST, and the TT) are placed in a threedimensional museum space, how these semiotics may have an impact on how visitors to move around in the museum space and to construct the "in-place" meaning of the museum. By bringing in the component of social actors, interaction order, visual semiotics and place semiotics, this theory proves to be useful in exploring the construction of social meanings in concrete and physical space. However, as pointed out above, geosemiotics has been proposed as a general theory, so research that intends to explore a particular type of physical space will need to substantiate the analysis of place semiotics by considering the specific features of that space -for example, by drawing from the studies of museum space in our case study.
In conclusion, it is hoped that this study has demonstrated how (non)provision of translations can potentially have an impact on the construction of in-place meaning in the multimodal museum space.