Professor Jerry Coyne: Polynesian navigation and Indigenous “ways of knowing”

One of the most-cited empirical achievements of indigenous peoples is the ability of Polynesians, Melanesians, and Micronesians to navigate their way across vast, unmarked expanses of the Pacific Ocean to get to small, uninhabited islands. How did they do it? And, regardless of their navigational methods, how many attempted voyages failed, with the navigators and their human cargo perishing? We’ll never know because there was no way to keep records.

One could think of the process of developing the techniques of cross-Pacific navigation as a form of natural selection on cultural variation, with the methods being refined incrementally because navigators who were successful survived to pass on their techniques. Over time, navigation would become increasingly refined and successful, though, even when “perfected,” people would still fail and die.

Finally, how did they know there were unihabited islands to go to, and where, roughly, they were? How, for example, did they get to Hawaii?

An article by science writer Patrick Whittle in Quillette, which you can access below or read part of it in this archived version, says that the methods of Pacific way-finding were uncovered by New Zealand physician and adventurer David Lewis by interviewing the few older Pacific people who recalled navigational techiques. His results were summarized in the 1972 book We, the Navigators, The Ancient Art of Landfinding in the Pacific.

Click on the cover image to go to the Amazon site:



I really need to read this book since the history and methods of long-distance voyaging have always fascinated and puzzled me. My most pressing question is “How did the navigators know where they were going; that is, how did they know there were nice uninhabited islands out there, and where they were?” I discovered that our library has this book (the second edition) and I’ve just requested it be pulled so I can pick it up.

Click below to read Whittle’s piece in Quillette (if you have a subscription)


Click to view

The point of Whittle’s article, though, is not navigation: it’s whether the methods of navigation qualify as indigenous “ways of knowing,” and how this relates to modern science. This is a topic we’ve debated at length, and I have to say that Whittle comes to a fairly sensible conclusion, one that I wouldn’t deny. And that is this: indigenous ways of knowing that have an empirical basis, like navigation, can sit side by side with modern science, and can be taken as science insofar as they express empirical “truths” (in this case, a set of methods that with some reliability can get you from island to island). But, says Whittle, indigenous knowledge like this oten comes with a cultural overlay: stuff like taboos, spiritual interpretations, and so on. Here’s Whittle’s thesis:

Across the Anglosphere, traditional Indigenous knowledge systems are increasingly being discussed not simply as accumulated practical expertise, but as distinct “ways of knowing”—often described as holistic and relational, and as inseparable from ancestry, place and lived experience, rather than, like modern science, a detached body of objective knowledge. These ideas form part of a wider movement to “decolonise” knowledge by challenging the historical dominance of Western intellectual traditions. In New Zealand, for example—the largest landmass settled by ancient Pacific sailors—the incorporation of Mātauranga Māori (traditional Māori knowledge) into the school curriculum has prompted fierce debate about the relationship between traditional knowledge and empirical science. At stake is not merely cultural authority, but whether different knowledge traditions should be judged by the same standards, and even what it means for something to count as knowledge in the first place.

In my view, the spiritual overlay belongs in the realm of anthropology and sociology, while the empirically true and replicable kernel of knowledge should be incorporated into modern science. That is not the argument Whittle mentions, though. The “fierce” argument, as in New Zealand, is whether the whole panoply of “ways of knowing”, including their spiritual, religious, cultural, and ideological components, should be taught as coequal to modern science, as in the entire worldview of Mātauranga Māori (“MM”; see my many posts on this subject here). It has been proposed that the whole gemisch of MM should be taught in New Zealand university classes, though I think that scheme is dead.

There are, however, two distinctions between modern science and indigenous “knowledge”. The first, which Whittle doesn’t mention, is that indigenous “knowledge’ is often scientifically wrong, but is retained anyway. One example is the NZ government-funded project to stem the fungal blight destroying kauri trees by dousing the trees with whale oil and playing whale songs to them. This is based on a Māori myth that kauri trees (Agathis australis) and whales were once related (brothers) in ancient times, but whales were sent to the sea. As the story goes, playing whale songs to the trees and dousing them with whale oil will enable whales to cure their landlocked and afflicted brothers.

This is palpable nonsense, of course, and though it’s part of the Mātauranga Māori “way of knowing,” there’s no scientific basis to pursue this project, and the Kiwi government is wasting its money by funding it.

Further, a huge part of indigenous “ways of knowing” involve spiritual and herbal cures for diseases, most of which have not been tested in modern scientific ways to see if they’re effective. Thus, any untested claims are not “modern science” in any sense, but hypotheses.

Second, the fundamental methods of establishing indigenous knowledge and modern science differ, with the former involving trial and error (as in navigation) and the latter involving hypothesis testing, pervasive doubt, toosl like statistics and blind testing, and establishing universal truths. A lot of indigenous knowledge is locality specific, like “here is how we create mussel beds in New Zealand.” That still counts as knowledge, but is not universal in the modern scientific sense. And the religious, teleological, spiritual, and mystical overtones of indigenous ways of knowing do not belong to modern science.

Like the seven Auckland Unversity professors who started off this controversy in 2021 with a letter in The Listener, I think that indigenous knowledge systems as a whole should be taught in anthropology and sociology classes, not in science classes, though the empirically true kernel of knowledge in such systems can fit into science class. (But I have to add that that kernel is tiny compared with the huge ambit of modern chemistry, physics, and biology.) So I don’t take much issue with Whittle’s conclusion:

Ironically, in trying to escape the seeming Western hegemony over what counts as “knowledge,” scholars sometimes simply replace one highly abstract category with another—Indigenous “ways of knowing”—that perhaps tells us more about contemporary academia than about the historical traditions it purports to describe. In reality, traditional practitioners like Lewis’s ancient mariners were not extraordinary masters of esoteric knowledge; they were ordinary human beings who had mastered the extraordinary capacities of cumulative culture.

Ultimately, when we clear away the institutional power dynamics and the culture war crossfire, a simple principle emerges: Indigenous knowledge deserves neither romanticisation nor dismissal. These traditions should be studied seriously, criticised where necessary, and appreciated for what they truly are—not timeless, inaccessible mysteries, but a vital part of humanity’s shared and evolving intellectual inheritance.

Who could disagree with that? But what we can dismiss are myths and religious stories like the actual brotherhood of kauri trees and whales. That story has about as much credibility as saying the Moon is made of cheese; neither is science and neither deserves serious study since there is no substantive basis for the hypotheses.

This still leaves me with the mysteries about navigation discussed above. Apparently Lewis, in his studies, discovered that the navigators had a completely (to us) non-intuitive system of wayfinding, described like this:

For example, [Lewis] was initially baffled by the concept of etak, or “moving island.” As he came to realise, however, the old navigators weren’t treating the canoe as travelling across a static ocean; rather, they were seeing their vessel as fixed beneath the stars, with the islands and the sea moving past it (much as passengers on a moving train might imagine the landscape sliding backwards while they themselves remain stationary). To keep track of progress, therefore, the navigator would visualise a reference island off to one side—the etak—that would drift backwards under successive stars on the horizon. By noting which star this island was under at a given time, the navigator could work out how far the canoe had travelled and where it lay along its route. What at first appeared mystifying to Lewis gradually revealed itself as an ingenious mental framework—a “polydimensional system,” as he called it—for organising spatial information without maps or instruments.

I still don’t understand how the concept of a “moving island” would work, since it requires reference islands nearby, and often there weren’t any. Further, the crucial question of “how often did this system work rather than fail?” must remain unanswered, as we lack the necessary data. Thus, even with the amazing navigation of the Pacific, we don’t have a scientific answer to a crucial question: was the system infallible or not, and if not, how fallible was it?

I’m just going to have to read Lewis’s book. There’s a fair amount of information about ancient navigation methods (use of the sun, birds, and so on) in the Wikipedia article on Polynesian navigation, including their use of double-hulled canoes:

Polynesians made contact with nearly every island within the vast Polynesian Triangle, using outrigger canoes or double-hulled canoes. The double-hulled canoes were two large hulls, equal in length, and lashed side by side. The space between the paralleled canoes allowed for storage of food, hunting materials, and nets when embarking on long voyages

Here’s a modern replica of one, labeled, “Hōkūleʻa, a fibreglass-hulled replica of a Hawaiian double-hulled canoe sailing off Honolulu, 2009″:

Posted to Flickr, uploaded to Creative Commons Attribution 2.0 using Flickr upload bot on 27 September 2010, 07:17 by Kahuroa.

Professor Jerry Coyne is an American biologist known for his work on speciation and his commentary on intelligent design, a prolific scientist and author. This article was first published HERE