The Aboriginal Culture theme is one of the most visited sections of Culture Victoria. So, when we designed the browse content by location function, we thought carefully about the relationship our audience has with the content, and how we could appreciate this through the ways we represented place.
This post, based on a talk I gave at the 2012 National Digital Forum, reflects on ideas of place, and outlines our partnership with the Koorie Heritage Trust to map the stories in Culture Victoria to 38 Indigenous language regions in Victoria.
Our relationship to place is intensly personal and complex. For example...
How do you answer: Your Parents? The city where you were born? A cultural affiliation? The planet you inhabit? The answer probably changes along with the context of how it is asked, and who asks it.
The title of this post, Ways of Seeing is a nod to John Berger's television series about questioning habits and convention in order to discover something about ourselves. Here is Jimmy Little introducing his song Yorta Yorta man...
Jimmy and I have something in common: our mother’s are from the same region. But when asked, I say my mum is from Deniliquin. I refer to the single point of the town (markedwith a red pin on the map), even though she grew up on a farm 20km outside town, whereas Jimmy refers to an area that is: “West of Albury, a little south of Echuca.” (circledon the map).
Oxford English Dictionary definition of place:
Noun. A particular position, point, or area in space; a location.
Why do we respond differently? Why do I choose the first section of the definition “a particular point” whereas Jimmy chose “an area in space”?
I can’t say for sure, but I think it has something to do with the fact that we have very different histories. I see Australia in large chunks, the East Coast for instance, and referring to a single point within a large chunk provides specificity.
This is in contrast to Jimmy, who saw the country as a series of small regions defined by language and culture. At the moment, around 150 Indigenous language groups remain out of 350 to 750 languages and dialects pre 1770.
It’s interesting that Jimmy, after referring to his parent’s country, then pulls his focus out, saying, "But my home and my stage is Australia, which is your home, and that’s where I’ll always be.” A consummate performer, Jimmy finds communion with the audience through place, and he can achieve this easily because:
The purpose of Culture Victoria is to increase access to Victoria’s cultural collections; it shares beautiful and fascinating material. When we did the last refresh of the interface, we tried to encourage browsing and discoverability.
We also thought about how to provide different avenues into the content, different ways to understand it. One way is by grouping the stories under 7 themes, another is by mapping it to the Framework of Historical Themes, another is by linking related stories. Another, the focus of this post, is to search story content by location, to discover it through place.
At the time of the rebuild, we were working with the Koori Heritage Trust on Meerreeng-an Here is My Country. Meerreeng-an tells the story of Aboriginal people of Victoria through their artworks and their voices, following a circular story cycle.
Each artwork is accompanied by a story, noting the storyteller and their language group, eg. How the Murray River was Made, told by Irene Thomas from the Bangerang language group and featuring artwork by Ralph Nicholls:
Because Indigenous language groups are such strong identifiers for place, and because the Meerreeng-an project included a wide range of language groups, it seemed not only logical, but necessary, to map the content to Victorian place names and Indigenous language groups.
We used the Gazetteer of Australian Place Names to map the Victorian place names. The Gazetteer deals with places that are areas (eg. the suburb Richmond) by pinpointing the area at its centre. It's pragmatic, but it's not optimal, and if you’re familiar with Richmond you probably wouldn’t pin its heart as being Coppin Street. But, well, maybe you would... it is afterall, just ways of seeing!
Indigenous language groups are also areas, not points (although you can posit that "...a point is really just a very small bounding box” as Aaron Straup Cope does in This is a Story about Naming Things, which if you’re interested in the complexity of location technology is a great read). However, language groups in this context are not points, they’re areas, and we felt that there must be a better solution than single point representation. And so, the partnership with the Koori Heritage Trust was extended, and the thought exercise began of how to:
The first issue we came up against was...
Culture Victoria is funded by the Victorian Government. However, when Indigenous people view this corner of the country they tend to see it more broadly as South Eastern Australia, extending up into NSW and across into South Australia.
We explored the possibility of mapping language groups across South Eastern Australia, not only because it offered an alternative way to see the country and fuzz up the state boundaries, but because it offered a way to meet the mandate of the website to increase access to collections... just because we deal with Victorian collections doesn’t mean the collections relate only to Victoria. If there’s material relevant to the Barindji language group, or to Deniliquin (both of which are just over the border), it would be only a good thing if these items were discoverable through our mapping.
But in the end, as it would add complexity to the project, we made the hard decision to limit the scope to what we were able to achieve. We would include the 38 widely recognised language communities that overlap the Victorian border.
The second issue we came up against was:
We had to determine the boundaries of the language groups. Now determine has an unequivocal ring to it... To be equivocal, I'd posit that it’s almost certain that the borders of the language groups have morphed and shifted for thousands of years. So to draw lines around them at any point in time would not only be difficult, but probably contentious. When you add to this the loss of cultural information due to the massive disruption following colonisation, the task is an even more complex one.
So, we aimed to...
To be as inclusive as we could. This meant taking into account multiple key sources to determine the language boundaries, specifically:
It’s interesting that through time these maps appear to get less complex - become less rather than more detailed. But what we have, in fact, in the less detailed maps, is a more accurate representation of a nuanced view.
It was with this in mind that we decided to NOT show visible borders in our interface. This would be a simple and effective way for us to acknowledge ambiguity and to achieve the goal of being representative but not definitive. And so, the text at the top of the right hand column specifies that content “is associated with Indigenous language groups that overlap the area shown on the map.”
But whether the borders were visible or not, we still had to determine them, otherwise the computer wouldn’t know how to relate the story content to a particular area.
To do this, Kent used the 3 source maps to draw the broadest, most inclusive, shapes for the language groups. Then, where possible, he checked his shapes with information and maps on the Registered Aboriginal Parties websites to include their interpretation of language group boundaries.
The Registered Aboriginal Parties view of the boundaries sometimes related very specifically to geological features, such as a creek or a dirt track, or where the Murray Rivers enters Lake Mulwala...
When the shapes were settled, Kent identified and plotted latitude and longitude coordinates to create polygons that matched the shapes as closely as possible.
To do this he used a version of David Horton’s AIATSIS map that had been overlaid on Google Earth. This enabled him to locate a lot of points relatively easily. But for others, where the area we were drawing differed from the AIATSIS version, or for areas that didn’t have a physical geographical feature identified by Google Earth, he had to go by eye.
He then sent the longitude and latitude points to Simon Sherrin, our Technical Manager at the time, who created visual polygons.
For example, this is the polygon for Yorta Yorta, Jimmy Little’s mother’s country:
And this is the polygon for Pangerang. You can see that they overlap quite a lot. Others overlap only slightly, or not at all. There is no rule.
Some were simple shapes to plot, like Yorta Yorta, others were quite complex, like the Jardwadjalirequiring around 50 points:
Kent checked accuracy by comparing the polygons against the shapes he’d drawn. Some were spot on, others required further refinement through additional latitude and longitude points.
The final element we had to consider was representation of alternative spellings of the language group names.
Whilst there are no right or wrong spellings, some people identify strongly with a particular spelling and so we needed the system to be able to deal with any one of these versions that might be used in a story. So, the spelling variations were added to the Language Area database tables.
So, where to from here:
At the moment you can only search by location, not by Indigenous language group. We’ll look at adding that functionality. Also, as the language group boundaries may change, we should schedule a review of the polygons at some point.
Now, I’d like to return to the question at the beginning of this post:
Interestingly, two people I’ve quoted in this talk answer similarly:
This is because there are many elements that make us us, there are many lenses through which to view and answer the question. The reason we did the mapping exercise at all was because we had something to attach to the places; something that gives the places meaning, the stories of how we interpret our world...
For example, The Aboriginal Object Collection at Dunkeld Museum shares changing perceptions of Museum practice and traditional owners. Possum Skin Cloaks includes elders talking about transformation through cultural practice. Interestingly, the possum skins used to create the cloaks are from New Zealand because it’s illegal to kill possums in Australia... In another example of trans-Tasman interaction, here's a video of Jimmy Little singing a cover of Into Temptation, a Crowded House song (who we Australians like to call our own, but we all know that’s not the only way to see it!):
These interactions and exchanges are a small example of how culture morphs and changes. The 'browse content by location' function is also a story in and of itself. By representing, side by side, two ways of perceiving and understanding place, it presents a third layer to the view.