Hello, everyone.
I'm Stacy Allison-Cassin.
I'm a librarian at York University
in Toronto, Canada.
I'm really happy to be here
with you all today.
I'll just also say right off the top
that I'm a citizen
of the Métis Nation of Ontario,
which is a recognized
Indigenous nation in Canada.
Also, the talk today,
I'm not going to show a lot of cool tools
or things like that.
It is about a conversation,
sort of picking up on the talk
that we just heard--
around issues, around Indigenous content,
Indigenous people, Indigenous culture
in Wikidata and Wikibase.
And I really want us to think about
a couple of key issues.
One is the relationship
between the data structures
we create and maintain,
and issues related
to human rights and equity.
So we should think--
We talk a lot about equity
in terms of gaps and accessibility,
but there are other ways
that we can also think about equity
in our projects.
So the ways that we can use Wikidata
as a space for activism,
making the world better for more people.
And modeling is hard, yet fun.
So I want to talk about modeling--
hopefully you want to talk
about modeling--a little bit,
and sort of invite you
into this conversation.
I think we are going to hold
some of the questions to the end.
But I also want to acknowledge
that what I'm talking about today
is not just my own thoughts,
that this is really building
on meetings like this, where we get
to talk together about things.
In particular, I want to call out
the Canadian Federation
of Library Association's
Indigenous Matters Joint Working Group
on Subject Headings and Classifications
that is doing work intensely
on this project right now;
also the National Indigenous Knowledges
and Language Alliance
Data Modeling Subgroup;
and specifically Camille Callison,
who's from the Tahltan Nation,
at University of Manitoba;
Dean Seeman at the University of Victoria;
Tim Knight, who's with me
at York University;
and Alissa Cherry,
who's at the Museum for Anthropology
at the University of British Columbia.
So what I want to center my talk today on
is this idea of sovereignty and nationhood
in relation to Indigenous peoples
and how this impacts
how we think about our data models.
So I want to talk a little bit
about colonization.
So for those of us
who are from colonized nations,
we have particular ways
that we might think about
how colonization impacts peoples and land.
So we know that
one of the goals of colonization
is actually to remove
the occupants of the land
so that people can come
and either settle that land,
they can engage
in resource extraction activities,
they're opening up the land
for agriculture
and other forms of habitation.
And we know that this is not in the past,
this is ongoing today.
We know that there are active things
happening in the world right now
which are seeking
to remove occupants from land,
sometimes lands they've occupied
for thousands of years,
in order to engage in these activities.
And we know that colonial states
engage in activities
to assure their control over territory.
And we heard a little bit about this
this morning in relation to language.
So we know that languages
aren't endangered
just through natural causes,
that there are deliberate actions
taken by governments or nation states
to eradicate language
in very deliberate ways,
because language is connected
to sovereignty.
It's connected to saying
that there is a culture
and people are active in this culture
and occupying the space.
When we think about how
small languages come to be small--
So I might get a little emotional
about these issues,
but these are the kinds of things
that are really important--
culture deliberately being eradicated,
and people, colonial nations
involved in acts of genocide
in various kinds of ways.
So that's a very serious topic,
but it does actually impact
the kind of work that we do
and I think is a thread that runs through
how we think about
the importance of culture
and the way that
dominant culture is deployed
within all kinds of cultural institutions.
So a couple examples from Canada.
There's many I could name,
but I'll just name a couple.
So currently there are fights taking place
in the province of British Columbia,
where the government of Canada
and corporations are trying
to build pipelines
through Indigenous territory,
and the hereditary chiefs
of the Wet'suwet'en Nation
does not want the pipeline built
through their territory,
but the government is actually
arresting people who are protesting,
even though they're on their land.
The Indian Act in Canada was instituted
as a deliberate way
to engage in assimilation,
so state-based assimilation tactics.
This is again through the removal
of language and culture.
Ceremony was outlawed,
so practicing your traditional ceremonies.
The traditional governance structures
for First Nations was outlawed.
A pass system was introduced,
so people were not allowed
to leave their reserves without a pass.
So you think about all the ways
that those methods
or the sovereignty of a nation
being actively worked against
and, again, these tactics of assimilation.
And then, of course,
many people here might know about
the residential school system in Canada,
which was children
being sent to boarding schools
where it was, again,
deliberate acts of assimilation
where you were stripped
of your language, of your clothing,
not allowed contact with your families,
and that's very deliberate.
So, again, going back to the keynote
we heard this morning
about parents choosing
to pass on their language,
well that choice is taken away
when children are sent away to school.
So that has long-lasting
intergenerational impacts
on the ways that families work
and on culture.
So the Truth and Reconciliation Commission
came out with calls to action,
some of those deliberately calling out
cultural heritage and education
in some of these places
where colonial practices
are actually ongoing.
So recognizing the right to self-govern
and to autonomy and sovereignty over land,
that's very important.
So it also governs our relationships
between nations.
We might use in Canada,
quite often, this idea
of nation-to-nation relationship.
So that recognizes that
the government of Canada is a nation,
but within Canada
there are also multiple nations.
So when we have a First Nation engaging
in negotiation with a government
that's considered
a nation-to-nation relationship.
Many Indigenous people in Canada
do not recognize Canadian citizenship.
They do not want to be associated
with being Canadian.
I know if I write Wikipedia articles
about Indigenous folks that I know,
one thing I have heard repeatedly is,
"Do not say I'm from Canada."
"I don't want to be
'so and so is an artist in Canada,'
or 'is a Canadian'... no."
So what does it mean
when we take that person
and we have a Wikidata item for them
and we say that
their citizenship is Canadian?
You know, that's actually
an act of violence against that--
I mean, it sounds very serious, but it is,
because we are saying that person,
who is actively working to resist
the colonial system,
and then we are saying in their data,
"Oh, but they're Canadian."
"Well, I want to be able to run
a SPARQL query against them
and bring up all the Canadians."
Well, that would be useful,
but what does it mean
when we replicate
these kinds of things in our data?
So recognizing Indigenous sovereignty
is an important aspect
in creating a more just
and equitable world,
even though we might not get
the kinds of data that we might want.
So if we're going to take
the strategic areas
of knowledge equity seriously,
we also need to pay attention
to the structures in our data.
So again, we tend to think along gaps
like the gender gap,
visibility gaps, small language
and marginalized communities,
but when we think about
why are these communities small,
or what does it mean
when we have these gaps?
And we have to, again,
think about the structures
and how we're conceptualized in our data
and how we're treating...
Just like the example of the photograph.
Again, why is that so bothersome
to the Sámi community?
It's because, yet again,
culture being appropriated,
them being misnamed.
Or again, and we see in Canada,
a return to the original
traditional names of territory.
And so all of these things
are really important,
and we have to think about
how we can center these practices
in the work that we're doing.
So again, I just want to emphasize
that belonging to a nation
is not the same thing
as belonging to an ethnicity.
I know sometimes that we think
about those things as being the same,
but they're not.
So again, it's thinking
about the relationship
between nationhood and nationality,
belonging to a nation and citizenship,
and the governant structure
that goes with that is different
than the ways we think about ethnicity.
And again, just to stress again
that it then becomes a conversation
around relationships between nations,
governance, land, and people.
So if we think about colonization
as an act of removing people
from their land,
or reducing their sovereignty
over the territory they occupy,
how can we, in the data that we produce,
recognize that these nations
are occupying a particular spot.
If we aren't talking about nationhood
and we talk about a territory
then we make those people
absent from that territory,
whether they're presently there or not.
So again, another thing to think about
is how we document occupation
over time, as well,
because one of the things
that you hear about,
especially in reference
to places like North America,
is that, "Well, no one was there.
It was a vast wilderness of unoccupied..."
Well, that's not true.
People have been living in North America
for thousands of years.
I have ancestors who have been living
in Canada, or the area of Canada,
for thousands of years.
So it's not an unoccupied space
that people just came in and discovered.
So this concept of discovery is helpful
in the ways that we think about
the colonial practices.
So I'm going to talk
a little bit about myself,
because I feel like I can.
Again, it's about being respectful.
I don't want to talk about
someone else's nations,
I'm going to talk about
my own a little bit.
So this is a picture of me and my dad.
So my grandmother,
my dad's mum, is a Métis.
And, just again, in reference
to the conversation this morning,
she did not teach her language to my dad.
She was living away from her community,
and it was definitely a thing
where you were not...
She did not want to talk about
being Indigenous.
That was not a safe thing to be
in the community that she was in.
I'm from Thunder Bay, Ontario.
I don't know how many people
we have here... probably not.
Anyway, it's not known--
It's known for having
pretty serious problems with racism.
And so that was her choice.
This is a picture of...
The young man standing in the back there
is my great grandfather.
And the document on the far side there,
I just want to talk about the ways that...
So you have this problem of Indigeneity,
or these kinds of culture
being suppressed in various ways.
But in the process of culture recovery
or in resurgence
or wanting to be connected
with a particular nation,
sometimes that becomes
a form of documentation.
So how do you prove you're a member?
There's saying you have
connections to the community,
but a lot of that
is through documentation.
This document here
is The Métis Petition of 1840,
from the Penetanguishene area,
and it's around when treaties
were being signed in that area.
Settlers were starting to come in,
they wanted the land,
so they had to have a treaty
so they could move
all the Indigenous people,
First Nations people, to an area
to free up the land for settlers.
That's a very crude way
of talking about it.
And this document is actually signed
by some of my ancestors.
It's a letter to the Lieutenant-Governor
at the time, saying, "Wait a minute."
Because it's called
The Half-Breed Petition.
So they're saying, "Wait a minute.
We are native also.
We should be included."
Because they called it "Indian presence";
they wanted to be included
in the negotiations that were going on.
So this became
a very important document presently,
in showing that this community
was expressing an Indigenous identity.
Because the Métis were not recognized
by the government
as an Indigenous people
until fairly recently.
So all of this is about being
outside of those negotiations.
And so one thing about this document
is it's in a collection,
a digital collection.
It took me forever to find it,
because it's just a scan of a microfiche.
So it was just like a...
There's nothing, there's no way--
So this is this super-important document,
lots of people want to see it,
and there's no metadata
in this collection that connects--
There's actually just zero metadata.
It's just like a long roll of things
related to correspondence
related to the British government
at that time.
So when we think about how also
we can surface documents
in a particular way
that are important to recognizing,
again, the existence
of Indigenous people in particular areas
is another thing
that becomes really important.
So again, this is talking about
my own nation.
When we talk about ways that we might
conceptualize nationhood or territories,
this is actually a map
of what the Métis Nation of Ontario
has designated as harvesting territory.
So that's actually related
to hunting and fishing rights,
and that was negotiated between
the government of Ontario
and the Métis Nation of Ontario.
The Captains of the Hunt are the people
who oversee that all of these activities.
So although I live in Toronto,
which is actually down here,
this would be considered
my traditional harvesting territory
because that's where
I can tie my ancestors to.
So when we think about how we might
model that kind of thing,
when we're thinking, again,
about structures in our data,
we need to recognize community roles
that also have ties to territory.
And then I'm going to talk
a little bit about--
Again, you don't have--
Not all Indigenous peoples agree
on what is someone's territory.
So there are disputes
between different things.
So recently the Métis National Council
has decided that this is the map
of the Métis Nation in Canada.
This does not recognize
the Métis people in British Columbia
or in some parts of Ontario,
so these other places are saying,
"Wait a minute.
We don't agree with this map."
So one of the things is who decides
or how are we going to negotiate between--
Is it actually allowing
for multiplicity of...
And then the First Nations people
whose land, this territory, covers,
were like, "Well, you didn't ask us
about this map."
So there is also thinking about the ways
that we need to negotiate
between claims on territory,
how we might document those claims,
but also allowing for recognition
that there is overlapping,
kinds of ways that we consider territory.
So I just wanted to post this quote,
because I think it's a really good way
of talking about how colonization,
we don't notice it, because it is,
in many places, the dominant culture.
It's the dominant way
we think about the world.
We don't necessarily notice
these kinds of things.
So again, when we think about
the perspectives of the marginalized,
so again, when we're talking
with all of us,
when we think about our data models
and our data structures,
how do we allow for properties or items
that maybe we don't think are important
but are actually vitally important
for all kinds of marginalized communities?
And this goes beyond
Indigenous communities.
This speaks to all kinds
of marginalized people.
And so we have to think about the ways
that we can use our data structures
to address some of these issues
and to become a space
where we actually are working for justice
within our data structures.
Okay. I don't know how I'm doing for time.
I forgot to put my timer on.
Oh my goodness! Okay!
(laughing) I've just got five minutes!
So I'm going
to speed through some examples.
Now I do have real-life examples.
I'm working with, as part of a member
of the CFLA Indigenous Matters Group
and NIKLA, we're working
on the development
of a First Nations, Métis,
and Inuit ontology.
We have developed this list,
this is just a little sample,
of all the kinds of things
that we're collecting
of what we're calling community names.
We had a soft launch
of this data on June 21st
for National Indigenous Peoples Day
and we are hopefully
going to be deploying this
within Wikibase, is the plan.
We have some stuff in there,
but I think we're going to have
to just wipe it and start over
because we're really--
We've come to realize where a part
of our work really lies
is in the data modeling.
So we really need to be thinking
about our data structures
and how we are going
to conceptualize that data
within the Wikibase environment.
Part of this is also related to Wikidata.
So I've kind of been ignoring
some parts of Wikidata
because I kind of don't want
to have to deal with some of it.
(chuckling) I'll just be really honest.
So there is "nation" in Wikidata.
So one of the questions I would have
is if you have Ojibwe, is that an ethnic--
Currently I think it's an ethnic group--
Is it an ethnic group? Is it a nation?
Is it both those things?
Do we have both those things
at the same time?
I think that's a question I have not yet
figured out how to answer.
We do have something called
"native land" in Wikidata.
When I first looked at it
a couple days ago, maybe last week,
I kind of stumbled on it,
and it actually was an instance
of an isolated human settlement
so maybe not the best way to describe
something that is called "native land."
So again, when we're thinking about
maybe it's good to check in with somebody.
I want to just show this example
of "Anishinaabe."
So Anishinaabe, here it's an ethnic group.
I would say it's also a nation.
But it's a nation
that also contains other nations,
so it's actually based on
kind of a language group,
but contains the nations
of Ojibwe, Ottawa,
and a number of other groups within that.
So how do we think about,
I don't want to say hierarchy,
but there's a way of--
A relationship has to be designated there.
Also, one of the things in that item
is a link to the official website
for the Anishinabek Nation.
The ethnic group
doesn't have an official website,
so do we have Anishinabek Nation
as an organization,
and then we have a nation,
and then we have...?
So there's a lot
of modeling questions that I have
around how we might want to work this out.
This is another example
of an archival item.
In the item record for this document
there's no actual reference
to Cherokee peoples
or how this information was collected,
so we might want to think about
how we relate some of these documents,
especially when they come
from a colonial government,
how they are documented in Wikidata.
And I just wanted to close
with this quote,
which is this idea of solidarity--
how do we stand in solidarity
with all kinds of communities
in our larger community?
How do we recognize, again, these places
where we really need to be sensitive,
and also recognizing that
some of these issues,
for some communities,
are vitally important
and it really does matter
how someone is called
or how someone is conceptualized
within our data
because it does matter what you see,
but also how it impacts
the larger internet and world around us.
And I'll close with that. Thanks.
(applause)
(moderator) Thank you very much, Stacy.
I invite back all of our presenters.
So, if there are any questions,
we've got lots of time.
We made them cramp a bit
their presentation
in order to let you express
your opinions or your questions, etc.
Also, thank you, Jon, for your work.
Yeah, there's a question there.
(Dragan Espenschied)
Hi, I'm Dragan, from Rhizome,
and I wanted to ask Stacy
what is your view or experience
with data itself being kind of colonial?
Because sometimes
I have the impression that
especially the things that seem ambiguous
are reflected in data
with the most descriptions,
and the idea of data to remove ambiguity
is kind of something that has struck me.
And I wonder, especially when you see
these disputed territories--
maybe no one ever cared about it before,
but now you have to describe it
and then suddenly it becomes a problem.
So what is your...?
Yeah, I feel like this is my life.
So I was a cataloger, for--
I don't know if anybody else here is a--
As you know, I'm a librarian.
I worked as a cataloger.
You can never get it right,
it always feels like.
You're always going to be...
There's no right answer, in a way.
There's only attempts.
But I do think that one of the issues
is that all of our structures
that we work with are colonial
and express power in different ways.
So there's no way that we...
We can't really "decolonize,"
I will say, many of our systems,
because that's just the way they are.
We we think about museums,
or libraries, or even sets of data,
that it's built into the code
in some ways.
So where are points
for resistance and recognition
within some of those systems,
and how do we work to change,
make systemic change from the beginning,
when we think about ways
that we start off?
There's, I don't know, it's like a scale
of better and worse things.
But I think if we're operating
from a point of consultation, of respect,
of recognizing human rights
when we take those things into account,
and how can we push
our organizations to do better.
So one of the reasons that we started
with having this ontology
is because it's actually to replace
Library of Congress terminology
in our libraries,
because in Canada we often use
Library of Congress terms.
Those terms were developed
for Congress in the United States.
They often don't fit
the Canadian experience.
Like the heading for Indigenous people,
for First Nations people,
is "Indians of North America" still.
And we have little hope that
the government of the United States
is really vested in changing those terms.
(laughter)
So it's part of--
Given that, what can we do?
And it is to develop our own ontology
that people can use
to replace those terms.
So I don't know if that's a great answer,
but I think there isn't--
We're always in those structures,
so what can we do
at various kinds of points?
(woman) I have a question for all of you.
How do you deal with pushback
when someone might say,
"Well, this is the answer
in a Western peer-reviewed journal.
This is how they called the people of 1890
and you're saying that this is inaccurate,
but where do you have your proof
when here it is
in a Western peer-reviewed journal?"
How do you deal
with that kind of pushback?
(moderator) Who wants to go first?
(laughter)
(woman) I'm not sure
which one of us is less likely to talk.
(chuckling) Yeah.
This is a horrible question.
Actually, it's a wonderful question
at the same time.
But, for instance,
if you look at the Sámi...
I recommend each
and every one of you today
to go look at the Wikipedias
and look at the different ones
and see what they call the Sámi.
The Sámi call themselves Sámi--
sápmelaččat in Northern Sámi,
sápmelaččat in Skolt Sámi.
In the Spanish Wikipedia it's lapp,
(chuckles sarcastically)
and lapp is a horribly racist word.
And there is a huge discussion about this
in the Spanish and Catalan Wikipedias
about what you can say--
"Well, Sámi's not in our language."
And I know it's been used in--
and I used to live in Barcelona.
I know it's used in Catalan, "Sámi."
And the Wikipedias have decided
they're going to use
the racist word instead.
Because it's not in any
peer-reviewed article somewhere.
So...
Yes, so... (chuckles)
But, I mean, we have
this session here today,
and part of it is we invite the community
to think about these things
and how we can...
What do you think we should do?
Part of it is what is
the appropriate evidence?
If it's used in one peer-reviewed journal,
do we have to collect evidence
somewhere else?
How do we encourage the community
to think about their responsibility
in this space?
And it's maybe a long process,
but when things are--
I think that's something,
especially in Commons,
when we have images,
I know there are lots
for North America
that are really problematic
and people will say,
"Well, it's public domain."
So I think that's a really good...
I don't have a quick or easy answer.
(woman) We'll need
to talk about that, yeah.
Yeah, I would like to be
a little bit optimistic with Wikidata,
because, well, I like Wikidata.
I think that the perfect side of it
is that we can express different views.
We can display
the peer-reviewed terminology,
but we can contest it with other evidence.
So I think this is... Well, it relieves
the responsibility to the respondent,
but still it gives new opportunities.
[inaudible crosstalk]
Sorry.
Okay, let's try this one.
Sorry, sorry about that.
(man) So, two things
I was going to mention,
but one of them is the one you just said,
that for situations where,
like you spoke about
the Canadian citizenship problem,
that can certainly be entered as,
you could say, "He's Canadian,
claimed by the Canadian government,"
or whatever this is,
and have a different thing that says
his citizenship is something else,
or even unknown, or even no value
if we don't have a nation...
if the nation is not allowed
by Wikidata in there,
which is a different discussion
that I guess you probably
will have to have at some point.
So this is perfectly doable in that sense.
The person's probably
still going to be unhappy
that the Canadian citizenship
is listed at all,
but at least you can show them that...
So it's listed as not a universal truth,
but only as one of the possible opinions.
Another thing I wanted
to bring up for a moment
is something I was talking to Kimberli--
it was kind of run through
through the slides
because of the time concerns.
This part was easier in the sense that,
okay, if you have two different things
and you can put
the two things there and it's okay,
but what happens for cases
where the community does not want
this knowledge to be public at all?
- Ah yeah, the Indigenous one.
- (man) Yeah.
I think there is space for us
to work on privacy, sensitive data,
and identifying those
and finding out ways
to handle content that we find,
or the communities find, problematic.
It's a large discussion,
and it has a lot of legal aspects.
It has a lot of ethical aspects,
and it ties to copyright as well
and the ownership of the content.
So, well, lots of things
to say about that.
Yeah, and I will also say
that you just-- [inaudible]
Copyright regimes that
we are familiar with are colonial.
There's actually a huge friction
between copyright regimes
that are used in most countries
and traditional knowledge.
I think we have to maybe
be comfortable sometimes
with deleting content,
even that we say,
"Well, it's public domain."
Well, public domain,
it does not necessarily have a meaning
in an Indigenous community
or in certain communities.
So what does it mean when, again,
when we go back to this idea
of sovereignty
and recognizing human rights
when we say--
I was at a meeting that
the Canadian government was sponsoring
on copyright regime in Canada
and Indigenous knowledge,
and someone said--and it just really
has stayed with me since that meeting--
"Human rights before property rights."
Again, if we are taking human rights
as our prime motivator
and prime way that we're thinking,
then some of these other questions
become easier to answer,
because we have to value humans
in a way, all humans.
So we can't say
that their property rights,
or something like public domain,
should come before that.
And it's hard.
It's hard for many of us who are
all about access to things,
access to documents,
because it's against
what we feel like we should do.
But in some ways I think
that's the direction
for certain kinds of content,
because a lot of things were collected
by anthropologists, for example,
and some of those things--
books or photographs--
are now in "public domain,"
and uploaded into Commons.
(moderator) Okay, so our session is over.
There was one more question
from that gentleman from the back, but...
- (man) That's fine.
- (moderator) Okay, sorry.
I really apologize for this.
So, thank you.
We'll be back tomorrow, so.
We have a meetup tomorrow,
at 11:30, I think.
If you want to talk more
about Indigenous issues, come on out.
(moderator) So... yeah!
(applause)