Need a smart, sensitive approach for representing indigenous peoples' land relationships

[jeffreyameyer]

Style change requested

This is a really tricky question and will require some thought and trial and error. Please reach out to Tod Robbins and @rwelty1889 on Slack for consultation on this topic for anything that might need a rapid response. And, as with other daunting challenges, I recommend we work toward finding a fast answer and working to iterate on it quickly.

There are a multitude of ways of describing indigenous peoples' relationship to land, land ownership, and boundaries - the latter two concepts which are so laden with our Western centralized-authority way of thinking that it can be difficult to think in any other terms. (see: https://native-land.ca/about/why-it-matters/)

I'm thinking of having a few different categories of relationships to land, with the emphasis being "few" for the purposes of visual distinction, but defer to better sourced taxonomies: - Homelands - to include land inhabited by indigenous people in any variety of pursuits of living - habitation, hunting, migration, community, etc. This may have little relation to any sort of territorial claim enforced by might and is completely distinct from any concept of administrative rule. And, it may not even be related to continuous presence on that land. - Territory - this is slightly more definitive, as it is claimed by and enforced by might. Examples in the current US boundaries include Comancheria (1750-1850) and various conquests by the Haudenosaunee (1450-1650), as well as the Navajo people and the Cherokee Nation. For now, we have Comancheria loaded into OHM. It's a great (and challenging) example of indigenous territory for us to use in testing, as it crosses Mexican, Texas Republic, then US boundaries, before being swallowed up by the US. - Cessions - this will be related to external territorial acquisition by a nation state by force or other means - e.g. the Louisiana Purchase, the Gadsden Purchase, Seward's Folly, the Anschluss, or the invasion of the Crimea. [We need a separate ticket for those types of expansion] But, it will be slightly different in that it involves a nation that has already occupied the territory of indigenous people first recognizing the claim of the indigenous people to their land (! otherwise, why negotiate?), and then negotating/coercing them under duress into giving up their land. We will have a fairly interesting Cherokee Nation cessions map loaded in the near future. - Reservations - these are sort of like country borders negotiated under duress. Not sure how these should be treated, but I would think just like any other national border for the purposes of our maps. While I respect the OpenStreetMap community's efforts to map reservations (e.g. Navajo Nation), I think there's a lot left to be desired. For example, ignoring the name, this boundary isn't really a boundary of "aboriginal" lands. It's an administrative boundary defined by a treaty negotiated in a coercive arbitration.

Affected Tag or Tags

I'm not sure how these various categories boil down into tags or rendering, but, at a minimum, I do believe we should endeavor to encourage users to cite sources of research to define time periods and geographic areas for any of these categories.

I'm assuming first that these should all be defined by relations and some common attributes:

type=boundary
place=homeland, territory, cession, nation/country
indigenous=[name]

I'm not quite sure how we'd want to tag a reservation yet.

Visual representation

I think the visual representation of https://native-land.ca/'s map to be very interesting, but not of much help for us (aside from the very interesting boundaries), because we need to show other information in context with other cartographic information. We may want to think of how we'll be showing other special types of regions on the map - e.g. disputed territories and also whether we might have some base map showing natural features. Google Maps appears to take an understated approach - a white transparent layer for the reservation land, but the font treatment for the Navajo Nation certainly doesn't feel "national" or respectful, esp relative to Albuquerque.

[jeffreyameyer]

Here's a zoom=9 level view in OSM that highlights some of our challenges:

[danrademacher]

thanks for all this great background! So it seems to me we have a few levels of challenge here:

1. How are the various levels of Native homelands and boundaries present in the data in OHM, and what tagging schemes support those?
2. How do we represent those cartographically?

I think Question 1 is actually a lot harder than Question 2. I mean, figuring out the map treatments will be hard, but first determining what is being mapped is a lot more challenging.

My first thought on that front: How can we invite Native people to be part of this conversation at a very early stage? I think that's critical, while also recognizing that we need to be careful about asking for their labor without also delivering value back to them. I wonder if we could use some resources slated for GreenInfo to enlist their help.

It's possible that we'd want to first get some very basics in place, as table stakes leading up to their involvement. In that case, I would recommend something like this:

1. Symbolize as label nodes only to start the centroids of all the territories on nativelands.ca. Whether the borders are in the data or not, I'm not sure, but as a first pass, symbolize the names on the landscape to ensure those can appear quickly without getting stuck on layering issues. Also, these should be "eternal" features, or at least from ~10-14K years ago to present, since those homelands remain homelands.
2. At the other end, we bring in the readily available public domain boundaries of current reservations and other lands based on the official BIA data (called Land Area Representation), so we have the current boundaries of reservations in the US. Start date is harder on these -- there's nothing like that in the source data, so each one is probably a research project.

Another approach would be to set a test area then focus outreach work, data imports, and cartographic design on that place. I'd love to do this for California, where we have fully recognized tribes like the Yurok and Huppa, as well as many tribes that lack federal recognition but have very clear traditions and senses of their home territories. The Amah Mutsun have begun to reclaim some of their traditional lands through land trust acquisitions.

[matkoniecz]

Here's a zoom=9 level view in OSM that highlights some of our challenges:

odd stripped lake fill is for marking intermittent water bodies

[jeffreyameyer]

Finally following up on @danrademacher's last comments...

Regarding recommendation (1) -

• Here's an .osm file containing label points for all of Native Land Digital's lands. If no one has objections, I plan to upload these on 6/3/22.
• I'm not sure that these should indeed by "eternal" features, as we should show earliest known presence for any particular group. For example, the oldest Coast Miwok artifacts found to date may "only" be from 4000 years ago. That said, I agree we should put a fairly distant in history start_date. ... and, we need to get the time slider to go back further than -4000... ;)

For 2) Agreed - that may be a good thing to do after we get the points entered.

[jeffreyameyer]

I think this graphic of the Tartessos in Iberia includes some interesting concepts (that are at odds with the title that includes "in 500 BCE")

• "Core" area - not sure if this is density in just space or space and time
• "Max extent" - clearly, a max extent implies a specific point in time, but it's also an interesting demarcation and reference
• "Influence" - again, this is subjective & not time-specific, but I think it's a concept that may be worth considering as we think about this type of mapping

[1ec5]

• Here's an .osm file containing label points for all of Native Land Digital's lands. If no one has objections, I plan to upload these on 6/3/22.

These points are tagged place=indigenous land, with the space – should that be an underscore? Also, would this be a tag similar to place=region that’s ambiguous in terms of scope? There are indigenous settlements within some of these lands that would eventually get mapped too.

[jeffreyameyer]

@1ec5 - yes - there should be an underscore there & I'm not sure if that's even right. I agree that the geographic description might be similar to place=region, but I think there are other distinctions of region (overloading natural and political uses of the term) that seem less desirable than what we're going after. Also, those uses of region generally refer to fixed spatial areas (I believe - could be wrong), rather than those that might move over time.

[jeffreyameyer]

Ok... while the feedback window is still open for the native-land.ca points, I uploaded a few of the SF Bay Area territory shapes, which was illuminating.

Data-wise, they are boundary relations with admin_level=1, to get through the various vector tile filters & place=territory to a) not be a "country", b) be easier to find with Overpass (even though I think they're the only admin_level=1 objects we have, and c) be available for styling.

Subject-matter wise... I have questions about the source data. There are many distinctions across the Miwok and even within the Sierra Miwok that not have been distinctions among the Miwok. From the main Miwok entry in Wikipedia:

Anthropologists commonly divide the Miwok into four geographically and culturally diverse ethnic subgroups. These distinctions were not used among the Miwok before European contact.

That said, native-land.ca's links to source maps are AWESOME - I've linked to them through the "More Info" structure in the navigator:

Check it out: https://native-land.ca/maps/territories/northern-sierra-miwok/

[jeffreyameyer]

@vknoppkewetzel - any thoughts on this topic?

I'm wondering if some sort of a less boundary-like line, maybe just a curved label or some sort of light area pattern might be more appropriate (just the label seems like the better choice to me...)

[jeffreyameyer]

Wondering if we should merge this and #553 as thematically-related issues. @vknoppkewetzel @tsinn @1ec5 @danrademacher

[1ec5]

Couldn’t hurt to keep two issues around. There certainly have been uncontested indigenous territories at various times throughout history (and prehistory).

[jeffreyameyer]

There's some interesting commentary on the challenges of mapping historical uncertainty here: https://github.com/aourednik/historical-basemaps @vknoppkewetzel

[vknoppkewetzel]

Yes blurring is a common and effective use for uncertainty! Excellent examples. In some cases I would even generalize the boundary further/make more organic (depending on the boundary in question)

Like this one is still showing the same boundaries and just blurred them . Generalization + blur also helps highlight "uncertainty", especially if starting with datasets that have a lot of "squiggles" like coastline derived from current-day data, for example