First Steps towards a Partial Genealogy of the Compass Rose

Matthew Edney

I had meant this to be a simple blog post suggesting how and why compass roses became a feature of early modern regional maps. It’s a rather small realization to which I came some time ago, but it has implications so I thought I’d share it here. Yet, once again, I found myself all tangled up in another detail of mapping and map history that has been understudied and taken for granted, so that it is rife with error and confusion.

Let’s start with a fundamental question: what is a compass rose? A search of the literature and the internet reveals a mess of images ranging from simple north arrows,



via detailed protractors, as found on modern charts, reading to single degrees from both true (outer) and magnetic (inner) north,]]

to the more traditional compass rose found on early charts and maps, which nowadays appear as some kind of talisman of authority and knowledge on sketchy, popular maps:

What these different graphic elements have in common is that all indicate direction, albeit with varying degrees of precision (with between 1 and 360 indicated points). They all have this functional core—the “compass”—which is wrapped up in a decorative and stylistic shell, the “rose,” which has the tendency to be both round and decorative (see OED “rose n.13”).

Each aspect has been understood in particular ways by historians. The decorative aspect supports occasional commentaries on the changing style of specific elements, such as when fifteenth-century Portuguese mariners added a fleur-de-lis to the north-pointing index. From this perspective, the compass rose joins the title cartouche as a map element that can be graphically manipulated and decorated without affecting the map’s function and utility. In this respect, the style can connote values of aesthetic elegance or commercial worth. However, the sheer variety of forms and styles over time defeats any coherent genealogy. Map historians have little option other than to accept each compass rose as a unique thing. By default, the decorative aspect has no history.

The functional core is consistent and universal. Any device that indicates direction is thus generically understood to be a “compass rose”; their semiotic value is necessarily only to denote direction(s). From this perspective, the compass rose has a history, but only in terms of how changes in the differentiation of directions has influenced the structure of the compass rose.

The principal moment of change occurred on medieval marine maps, when the wind diagrams integrated within their networks of wind lines were converted to emulate compass cards after the introduction of the magnetic compass to Southwest Asia and Europe in about 1300. The earliest known instance of the altered device was on the ornate world atlas of 1375 by Abraham Cresques (1325–1387), the so-called Catalan Atlas (Wallis and Robinson 1987, 245–46). (This transition has given rise to no small confusion around the term “wind rose.”) As the parallel to the magnetic compass was accepted, wind lines were reconceptualized as “rhumb lines,” a rhumb apparently being a Portuguese-derived word for a point on a compass and therefore one of the primary directions. On the plane geometry of the medieval “portolan” chart and of the early modern “plane” chart, the two kinds of lines are coincident, but on a projected surface, as on charts made on Mercator’s projection, wind lines and rhumb lines (i.e., loxodromes) diverge and are not actually the same. (I said at the beginning that this topic was confusing.)

We might also consider the history of the appearance of compass roses on property maps, especially property maps but also urban, topographical, and chorographical maps, as a function of the adoption during the Renaissance of the magnetic compass in plane surveying. (This adoption was perhaps made independently in different places.) The compass rose per se seems to have been dropped in favor of north arrows in the eighteenth and nineteenth century as surveyors increasingly mapped with respect to true rather than magnetic north.

From this strictly functional perspective, the appearance of the compass rose on geographical maps requires some explanation. In particular, the maps’ meridians of longitude and parallels of latitude comprehensively defined the four cardinal directions.

In some cases, geographical maps were made to emulate marine maps. This was the case, for example, with Simon de Passe’s famous geographical portrait of John Smith and New England, first published in 1616|7:

Simon de Passe,  New England/John Smith , state 4 (1624). Leventhal Map Collection, Boston Public Library,]

Simon de Passe, New England/John Smith, state 4 (1624). Leventhal Map Collection, Boston Public Library,]

De Passe’s emulation of marine maps was inept (Edney 2010, 2011). To begin with, he used the geographical convention of stippling to denote the sea and to distinguish it clearly from the land; contemporary printed marine maps did not use such stippling. Contemporary charts were drawn with a network of wind/rhumb lines crossing the entire surface, land and sea both, but de Passe drew such a network only over the sea. More important, de Passe drew two networks, not one: the lines radiating out from the decorative compass rose should intersect with the two undecorated “wind roses,” but they do not. In other words, the map’s compass rose and wind/rhumb lines might have denoted direction but they also connoted marine-map-ness.

In considering this connotation, I realized that compass roses on early modern Dutch and English regional maps, if present, generally occurred as just one or two compass roses, perhaps with radiating wind/rhumb lines, set in and limited to the open sea. The compass roses are not placed in lakes. If present, the lines do not make a complex network but just one or two sprays, and the lines all end at the coastlines and do not extend over land.

Two compass roses set off the North American coast. Joannes de Laet,  Nova Anglia, Novum Belgium et Virginia  (Leiden, 1630);]

Two compass roses set off the North American coast. Joannes de Laet, Nova Anglia, Novum Belgium et Virginia (Leiden, 1630);]

My sense is that map historians have assumed that this feature was an inertial derivation from, and simplification of, the structural symbols on the sea charts from which the coastlines on these geographical maps were ultimately derived. Such an explanation seems to reinforce the modern cartographic ideal and its fundamental presumption that map data always stems from original observation and measurement. In this respect, the compass roses appear to have a strictly denotative character and as such they have not been included in the two remarkable studies of map signs on early modern maps (Dainville 1964; Delano Smith 2007).

But the sheer conventionality of the single-compass-rose(-and-radiating-lines) and their structural irrelevance to geographical maps based on latitude and longitude suggests that something more is going on. Specifically, it seems that the presence of compass roses (and wind lines) served as a geographical symbol for “ocean-sea”/“not land”. The symbol seems to have been a Dutch innovation in the sixteenth century that was subsequently copied (with many, many other aspects of geographical mapping) by the English. The innovation was also adopted by the French.

A very simple compass rose is set in off the coast of Labrador, amongst many other signs of open sea; a second compass rose, set to the north of Lake Superior, on land(!), seems to belong to the inset of the Gulf Coast, directly above it. Zacharias Châtelain,  Carte de la Nouvelle France  (Amsterdam, 1719);]

A very simple compass rose is set in off the coast of Labrador, amongst many other signs of open sea; a second compass rose, set to the north of Lake Superior, on land(!), seems to belong to the inset of the Gulf Coast, directly above it. Zacharias Châtelain, Carte de la Nouvelle France (Amsterdam, 1719);]

The realization that compass roses connote some further meaning in addition to denoting direction was not a revelation. After all, as I noted above, the compass rose is a key element on many modern popular and pseudo-antique maps, whether to connote structured knowledge or quaint antique-ness.

But this realization does point the way to how we might construct a genealogy of direction indicators on maps. We cannot continue to treat them all, with their variant forms and stylistic heterogeneity, as being essentially the same. We must first reverse our habitual lumping of all those indicators into a single category and instead discriminate between:

• north arrow, with or without the distinction between true and magnetic north

• cardinal cross: a cross with four points for the four cardinal directions

• wind rose, perhaps labeled with the names of the words or their abbreviations

• compass rose: 8, 16, 32–point figure emulating the card in a magnetic compass

• wind diagram: a complex of lines radiating from a point, their length and shape indicating the frequency and strength of the prevailing winds, as developed in the nineteenth century by Matthew Fountaine Maury

• compass diagram: the indication of 360° in one-degree increments on a modern sea chart

These terms are off the top of my head and I am not wedded to any of them. Alternative suggestions are welcome!

And we need to consider the occurrence of each within the relevant discourses, to establish not only their function use for the consumer but also their connotation of further signification within those discourses. In this way we might be able to trace how certain direction indicators were added to some kinds of geographical maps, but not others. In other words, we must remember that, like all other elements of a map, the direction indicator is just another representational strategy, subject to discursive conditions.


Dainville, François de. 1964. Le langage des géographes: Termes, signes, couleurs des cartes anciennes, 1500–1800. Paris: A. et J. Picard.

Delano Smith, Catherine. 2007. “Signs on Printed Topographical Maps, ca. 1470–ca. 1640.” In Cartography in the European Renaissance, edited by David Woodward, 528–90. Vol. 3 of The History of Cartography. Chicago: University of Chicago Press.

Edney, Matthew H. 2010. “Simon de Passe’s Cartographic Portrait of Captain John Smith and a New England (1616/7).” Word & Image 26, no. 2: 186–213.

———. 2011. “A Cautionary Historiography of ‘John Smith’s New England’.” Cartographica 46, no. 1: 1–27.

Wallis, Helen M., and Arthur H. Robinson, eds. 1987. Cartographical Innovations: An International Handbook of Mapping Terms to 1900. London: Map Collector Publications for the International Cartographic Association.