Here's a great example of non-Indigenous mapping making claims about time, space, collective identity, and political legitimacy! Obviously it would be inappropriate, and probably imprecise, to say that Willard is constructing a kin-space-time, but this example highlights for me how the Indigenous/non-Indigenous binary leads to something of an unhelpful caricature of the latter. Just because we might not think that Willard's claims—about history, legitimacy, or inevitability—are as convincing or legitimate as those of the peoples murdered, displaced, or erased by European settlement doesn't mean that her maps are not using some of the same techniques (with important differences, of course).

Two comments here.

One, I'm probably less comfortable with a category like "Indigenous mapping" than others (including, perhaps not coincidentally, some who identify as Indigenous). To me it seems to have both the plusses and minuses of pan-Indigenous politics more broadly, making connections and a collective voice but also flattening the diversity of Indigenous practices, cultures, and political goals. It also seems historically specific, emerging in the late twentieth century as part of the politics of the late twentieth (and early twenty-first) century; it's not a neutral description either of historical or contemporary practices. The idea that cultures and practices as diverse as Marshallese stick maps, the Relaciones Geográficas, and the oral history of communal hunting grounds should be unified as "Indigenous mapping" I think does too much to create a binary between Indigenous and Western/modern/etc., precisely in all the ways that this kind of binary is problematic. I think it's more powerful—both analytically and, beyond the specific goals of pan-Indigeneity, politically—to rather see different cultures and traditions as in fact quite different and not to collapse all differences into only two options (settler and Indigenous, modern and traditional, us and them, etc.).

Two, one of the main thrusts of recent scholarship on cartography is to argue (successfully, in my view) that all mapping, not just Indigenous mapping, should be analyzed as process. (Edney's recent "processual" approach is one example, but there are many others before him, going back at least twenty years.) I also think it's just incorrect to say that mainline mapping, even from the point of view of its own practitioners and institutions, is not interested in process, relationality, temporality, or human geographic experience. So much of mainline mapping is concerned with precisely these issues, even (especially?) in cases where maps are deployed as tools of management, control, and expropriation. There's also a lot of genuine diversity within mainline mapping—navigational charts, school atlases, and cadastral maps are all quite different in conception and in practice, and collapsing them into a monolithic whole can easily lead to errors of analysis. (Or, put another way, I see a contrast here between the care and nuance of your reading of Indigenous mapping with the somewhat overbroad and unevidenced characterizations of mainline mapping. There is no single "logic" of mainline mapping, just as there's no single logic of Indigenous mapping.)

Several of the map zoom-ins here aren't corresponding with the text.

Something about this line of argument in the chapter is making me uncomfortable, and I'm trying to put my finger on why. I think maybe there are three issues:

1. Is the ultimate implication that no knowledge should ever travel, that no one should ever attempt to analyze others' experience? If all attempts to make knowledge mobile, legible, and communicable are colonialist at their core, do we just give up and do nothing at all? My guess is that this is not what you mean, but perhaps there's a slippage here between how the map is made (through possible coercion, in service of further subjugation, without the consent or participation of those being mapped) and the representational logic of (some) mapping more broadly. I think we can argue against the former without collapsing it with the latter.

2. Is there an implication in this chapter than non-colonial knowledge (or subjugated peoples) would reject values like clarity, insight, or significance? This seems wrong to me on several levels; at best it seems insulting, at worst exoticizing and essentializing.

3. As much as I understand and respect the politics behind Indigenous data sovereignty, I'm not sure it follows that this should be the template for the treatment of all data. Not just because it seems to imply that no one should ever share knowledge or try to understand others' lives, but also because questions of sovereignty are not really at issue with most data. (If Belgium wants to restrict outsiders' access to the Belgian census I guess that's fine, but I don't think that's really what you're gesturing toward.)

Whoa! This is a bit of a headspin, right out of the gate. The idea of the "thematic map" is thoroughly an invention of the twentieth century (Max Eckert, Arthur Robinson, and friends), and projecting this idea backward in time seems like a way of folding earlier and competing ideas of space and representation into the norms of twentieth-century academic mapping (including its apolitical ideals, its divisions of labor between "concrete" and "abstract" knowledge, and its disavowal of its obvious connections to nationalism, military power, etc. [but less "capitalism" than one might suppose!]). I suppose you could do this as a way of trying to reclaim and rethink this category, but the more immediate impression is that of saying that Shanawdithit is worth taking seriously because she was no less modern or analytic than later mapmakers, which seems... off?

Throughout this chapter I was reminded of some of the arguments in The Imperial Map (2009). In this instance here I think of Hostetler's good argument against the two poles of the East/West dichotomy (Joseph Needham vs. Cordell Yee, both of which reproduce the binary even as they argue against its most offensive claims). More broadly I also think of Edney's argument that what might seem like a commonsense distinction between imperial and non-imperial mapping doesn't really work.

Suggesting that all data and all visualization is inseparable from racial capitalism is a strong claim, but in this intro the connection is made more at the conceptual level (quantification, abstraction, etc.) than at the kind of historical level that would be required to make this claim fully convincing. (My guess is that historically there's a lot of overlap between practices of data and visualization and the development of Euro-American capitalism, but that there are also ways in which data and visualization have not aligned with capitalism but instead with other, perhaps even competing, cultural-political projects. The same is true for the long history of quantification and abstraction themselves.) But even just staying at the conceptual level, do you mean "racial capitalism" to refer to a specific kind of capitalism, or to mean that all capitalism is racial? The first seems more coherent—historians would have us always refer to capitalisms, not any single universal capitalism—but also less convincing in the case of data and visualization, since there are plenty of practices of data and visualization that cannot be so easily folded into the history or politics of race, capitalism, or empire.

More broadly, I think that pitching this project as a close focus on how data and visualization intersect with capitalism, racism, and empire is compelling and coherent, but that the stronger claim—that all data and all visualization are always handmaidens of racial capitalist imperialism is harder to argue and ultimately weakens rather than strengthens the argument. In other words, bringing our attention to the intersection of data, visualization, race, capital, and empire seems stronger as something more specific rather than more comprehensive. (And indeed, shouldn't we be wary of all attempts at totalization and macrohistorical metanarrative?)

I also see something of a contradiction here: if data and visualization are inseparable from racial capitalism, then what do we do with artifacts like quipu or Indigenous mapping? Are these not data and not visualization (because they're not part of racial capitalism), or should we somehow see them participating in the same logic of commodification and abstraction as all other data and visualization? It's unclear how to expand of our frame to include non-Euro-American practices without also bringing more specificity to our understanding of the Euro-American mainline.

You've used "must" a few times by now, and I'm not sure why it's the right word. I can easily see how making (or reading) visualizations with an awareness of the historical and present-day politics of colonialism, class, and inequality could be good, even transformative, but I'm less convinced by the imperative here that all visualizations must be read or made this way. (For example, what's gained by reading or making visualizations of protein folding with a decolonial frame? If I keep track of my research spending with an Excel chart, do I need to see this in the service of empire?)

I say this not to suggest that the apolitical ideal should be defended, or even that neutrality is possible at all, but because "all data and all visualization is racist and colonial" actually seems less convincing, and less compelling, than something more like "many practices that might seem innocent or neutral—either by practioners or by their historians—are in fact anything but, and certain core assumptions of the field need to be revised as a result." (Or put more simply, if everything is racist and colonial, then nothing is?)

Red flag! Who is "we" here? And why wouldn't "we" want to believe this? I think there are plenty of people who would have no resistance to this claim (even without evidence), and others that would reject (even in the face of strong evidence). To avoid this being a rhetorical straw-man, I'd rather know if there is actually disagreement about this (either scholarly or otherwise) and what or who your specific target here is.

This seems like a core tension in the project: how much are you treating "data visualization" as a historically specific category, and how much are you trying to challenge precisely that category? The idea of a "counterhistory of data visualization" could mean several different things, some more coherent than others. This could be an attempt to reinterpret mainline data visualization (highlighting, say, the inescapable politics of nineteenth-century statistical conferences); it could be a challenge to the historical genealogies of mainline visualization (arguing for a different periodization or the overlooked importance of certain people or institutions in mainstream visualization); it could be a broader challenge to the coherence of "data" or "visualization" as historical phenomena (arguing that neither data nor visualization should actually be seen as rooted in Europe and its offshoots since the late 18th century, but perhaps more widely in global early modernity, for example); it could also be something more sprawling about all attempts throughout human history to communicate knowledge through images. Right now this project seems like something of a mix of all of these, and I think it could be helpful to give a more focused account of its goals, both specific and broad.

(I'm perhaps reminded of the famous line from Steven Shapin: "There was no such thing as the Scientific Revolution, and this is a book about it.")

(All these comments could also apply to other key categories in the book, especially "statistics." Does this just mean any numerical value, in any time and any culture, or does it refer more narrowly to statistics as a historically specific idea and practice?)

I'm reminded of John Pickles's article on propaganda maps from 1991, where he argues that "The interpreter must give an optimal reading of the text and of the meaning the text must have had for those for whom it was written and show what the text means for us today in the context of modern views, interests and prejudices." That is, I wonder if there's more you can say here about what you mean by "context" and the goals of contextualization. This can mean different things—some banal, some pedantic, some fussy and antiquarian, but also some profound and consequential. Is the goal to recover what an artifact meant at the time it was created (perhaps with different meaning for different actors), or to change the meaning of the artifact in the present in ways that historical actors might not have seen at all? Or is this simply a call to prioritize the synchronic connections of an artifact to other coeval artifacts and texts instead of prioritizing its diachronic connections to other artifacts in a longer (often Whiggish) genealogy? How would you explain the goals and methods of contextualization to someone who wasn't already a humanist?