Model of landscape evolution
by William Morris Davis



Date: Wed, 4 Nov 1998 09:15:54 -0700 (MST)
From: Paul R. Larson
Subject: Wm Morris Davis

I would like to poll the opinions of the GEOMORPH community. What is the current feeling about the model of landscape evolution proposed by William Morris Davis in the early part of this century (landscapes classified as being youthful, mature, or old age)? I have talked to some who no longer accept it, and would like to know how the list members feel.

Second question: If we no longer accept it, what has taken its place?

Thank you,

Paul R. Larson

William Morris Davis (1850-1934)

William Morris Davis

William M. Davis, famous American geographer, geologist and meteorologist, began his career in geography at age twenty, serving as a meteorlogist in Argentina. He held his first major teaching position at Harvard University from 1876 until 1912. He founded the science of geomorphology, the study of landforms. His major contribution to geography, though, came in the form of his Davisian system of landscape analysis, which involved recognizing the long-term, cyclical nature of erosion in landforms and landscape analysis. In 1904 he helped found the Association of American Geographers. By the time of his death, he had published over more than 500 works.

[Sources: Britannica Online, Encarta]

Some links:

  • William Morris Davis
  • Geographical Essays by William Morris Davis
  • William Morris Davis. A Biographical Look

    WWW Editor of The Association of Polish Geomorphologists

  • R E S P O N S E S

    Date: Wed, 4 Nov 1998 14:59:29 -0600
    From: Jack Shroder

    As one who ran aground on Davisian geomorphology in the late 1960's, let me tell you a story about it. In my Ph.D. dissertation, in the absence of very many quantitative measures of age (14C dates, etc.). I used various "aging" criteria to show crudely how landslides changed over time through erosion and soil development, etc. to lose their definition on the landscape. By using my criteria one could gain a better understanding of regional mass movement in order to pick out potentially unstable areas, providing that conditions changed (climate, development, etc.). I published and presented this at an international meeting in 1968, with criteria for recognizing youth, maturity, and old age. Gordon Wolman and Asher Schick got all over my case. I never mentioned it again, until 15 years later I received some nice mail from people who discovered the publication and put it to good use in understanding landslides elsewhere. So in getting past the dramatic over-applications and inappropriate applications of Davisian geomorph, we threw part of the landscape-aging baby out with the bath water.

    Schumm and Lichty's famous paper on time and scale in geomorph (AJS) is the best on the subject I have ever seen, because it is clear that landscapes must erode down over time. The question is how, and has a tremendous chronology-bound hook attached. Dynamic equilibrium and the new understandings of tectonism (uplift) and multiplicity of reasons for renewed downcutting ("rejuvenation"!) must be addressed. Peneplains probably don't exist anywhere, but polygenetic, polytemporal surfaces of erosion are everywhere on the landscape and in the rock record.

    What we replaced Davisian geomorph with is not the dramatically over simplified denudation chronologies of landform development, but rather the modern understanding of landform complexities, threholds, climate and process change through time, rock (lithologic) control of landforms, the convergence principles, complex responses, and many other ideas. Hard to teach at an intro level, but al least a bit less misleading to the uninitiated, one would think.

    Hope that helps.

    Jack Shroder

    Date: Wed, 04 Nov 1998 15:42:55 +0000
    From: Jim Rice jrice@LPL.Arizona.EDU

    I also thought that these terms were no longer accepted, however I have most recently come across them in a book entitled The Geology of Arizona by Dale Nations and Edmund Stump (1997, 2nd edition ). I would be interested in knowing both why this idea was abandoned (when was it decided and by whom) and what has taken its place.

    Jim Rice

    Date: Wed, 04 Nov 1998 18:59:28 -0600
    From: Gerardo Bocco V.

    I wrote a paper in 1991 (Gully erosion: processes and models, Progress in Physical Geography 15(4):392-406) where I discussed the davisian approach as far as gully erosion modeling is concerned. My view was quite critical. If anyone is interested, I would be happy to send a copy of it. Basically my opinion was based on my experience in accelerated erosion research in recent volcanic landscapes, very much influenced by spasmodic contribuitons of lava and pyroclastic materials, as well as tectonic uplifts. In these conditions, the cyclic model may not be the best approximation to landscape evolution. As far as I know, there is no model (at least of the same scope of the davisian) replacing Davis's one. I was taught, very long time ago, that the Walter Penck's model was an alternative to Davis. However, I think both consider gradual changes in the denudation chronology. In areas that are very active, because of both endogenic and exogenic processes, these models are not quite suitable. I don't know if recurring to chaos theory may be something more than a fashion.

    Cheers, Gerardo Bocco
    Gerardo Bocco.
    Investigador Titular Laboratorio de Geoecología Depto.
    Ecología de los Recursos Naturales Instituto de Ecología de la UNAM-Campus Morelia.
    Tel. 52 43 200517. Fax: 52 43 200830
    Dirección postal: AP 27-3. Xangari. 58089 Morelia, Mich. México

    Date: Wed, 04 Nov 1998 19:09:00 -0600
    From: Phillip R. Kemmerly

    I have not accepted it for the last twenty years. What has replaced is an entirely different focus on process-mechanism-landform or product.

    Phillip R. Kemmerly

    Date: Fri, 06 Nov 1998
    From: John Jansen []

    When is an hypothesis refuted?

    The dogged persistence of the Davisian model of universal down-wasting as well as the similarly inadequate Penckian model of scarp back-wasting are prime geologic examples of the importance of sociological factors in governing theory acceptance and rejection. Davis’ model owes a lot to Darwin and together with Marxism, the formulation and rise of these should perhaps be viewed in terms of Modernist demands for the ‘Big Universal Theory’. If Davisian cyclicity has limited value in the tectonically vigorous volcanic landscapes described by Gerardo Bocco, it is also a complete turkey when applied to stable cratonic and passive margin settings such as those found in Australia (and other Gondwana fragments); here the extensive preservation of Mesozoic and Palaeogene etch and/or exhumed surfaces high in the landscape poses a major challenge to peneplanation.

    A series of papers by Twidale (starting in the 1970s) dealing with long-term landscape development in cratonic settings has re-cast light upon the work of C.H. Crickmay and the continental workers Bourcart and Jessen. While Twidale’s work may not be particularly relevant to landscapes scraped clean by the Quaternary glaciations, such areas cover just a portion of the Earth’s surface. For a good cover of the literature checkout Twidale (1994) in Paleao 3, vol.112:157-186. Also of interest may be Paul Bishop’s Popperist critique of the Davisian model - I’ve lost track of that reference.

    John Jansen
    School of Earth Sciences
    Macquarie University Australia

    Date: Fri, 06 Nov 1998
    From: Dick Marston

    What has taken the place of Davis and other all-encompassing models of landscape evolution is best presented in Stan Schumm’s book, “To Interpret the Earth: Ten Ways to Go Wrong.”

    Dr. Richard A. Marston, Professor,
    AAG Secretary, Regional Councillor AAG Great Plains-Rocky Mt. Div.
    Department of Geography & Recreation University of Wyoming Laramie, WY 82071-3371
    DIRECT PHONE: 307-766-6386
    DEPARTMENT PHONE: 307-766-3311
    FAX: 307-766-3294

    Date: Fri, 06 Nov 1998
    From: David Amsbury

    Second question first: I don’t think anything has replaced it. There’s no GENERAL theory in any book or paper I’ve found that helps me understand the present shape of a landscape, nor interpret its history. The closest thing to a principle I’ve found is the idea that climate is continually changing at unpredictable rates in unpredictable ways with unpredictable effects. Tectonics, also: the jostling of plates is unpredictable in direction and rate. So what you see is the record (in sediments and erosional features) of what happened at that place to those materials during this period of history.

    First question: There’s no place I know in the world today where a long Davisian cycle can be demonstrated. Even in Texas, which has not experienced continental glaciation as far back in time as there are strata, volcanic activity since the Oligocene, nor serious “mountain-building” tectonics in most of the state since the Miocene or maybe the Oligocene, there’s no evidence for Davisian stages of maturity and old age. Erosion during the late Tertiary and Pleistocene has been episodic. Even though much of the (pedimented) landscape is stable and has been preserved by a mature caliche soil since the middle Pleistocene, there’s no progression toward “old age” or reduction of relief. Quite the contrary. Relief has increased, though episodically, since the late Miocene. Along the main southern road into Death Valley the slopes are mantled with the debris of a formerly much wetter climate than today’s; this is true of all the Southwestern US except for small patches of badlands adjusting to today’s climate. In southeastern Australia, “youthful” valleys are preserved by lava flows as old as Eocene; relief hasn’t been reduced there during the past 50 million years or so, it has increased. In the “Arid Heart” evidence of formerly wet climates (lake beds, deeply-incised stream valleys, mantled slopes, ancient deep soils) is everywhere interspersed with sand sheets. No Davisian cycle there. To borrow some pop psychology from paleontological popularizations, I suspect that the Davisian cycle and similar theories of landscape “development” were informed by a belief that the world has been guided to near-perfection just for us Victorians at the self-evident peak of intellectual evolution.

    David L. Amsbury 128 Homestead Kerrville, TX 78028

    Date: Fri 06 Nov 1998
    From: Richard A. Marston []

    In 1996 I presented a paper in Nancy, France, on the changing status of Davis’ Cycle of Erosion Model. This meeting was held on the 100th anniversary of the publication of Davis’ benchmark paper on river capture of the Seine, Meuse and Moselle rivers. My co-authprs, Jack Vitek, Paul Candelaria (grad student) and I surveyed 85 textbooks published in English (or translated into English from other languages) in the 20th Century. 51 of the texts were introductory texts in physical geography or physical geology; 34 were advanced textst in geomorphology or physiography. We did not include texts written by Davis.

    The treatment of the Davisian cycle of erosion concept was classified as follows:

    1. presented as a fundamental concept
    2. concept described but also criticized
    3. concept mentioned for historical interest only as a step in unifying the discipline
    4. no mention of Davis or cycle.
    Our results show a shift, tabulated decade by decade, from 1 to 4...although the pattern differs somewhat in timing for the two levels of textbooks. I am writing these results up as we speak for submission (finally) to a journal. I would be interested in any observations from others who have examined this topic, and would gladly give credit in the manuscript as appropriate.

    Dr. Richard A. Marston, Professor,
    AAG Secretary, Regional Councillor AAG Great Plains-Rocky Mt. Div.
    Department of Geography & Recreation
    University of Wyoming Laramie, WY 82071-3371
    DIRECT PHONE: 307-766-6386
    DEPARTMENT PHONE: 307-766-3311
    FAX: 307-766-3294

    Date: Mon, 09 Nov 1998 08:50:56 -0600
    From: Bruce Rhoads

    In response to the Paul Larson's inquiry, I will put in a shameless plug for the book The Scientific Nature of Geomorphology, 1996, edited by myself and Colin Thorn (Wiley). This volume was intended to provide an overview of philosophy and methodology about 100 years after Davis and Gilbert. In our paper, Observation in Geomorphology we argue that the problem with Davis' approach to geomorphology was that it was excessively theory-laden - to the point where Davis was never willing to explicitly identify the type of evidence needed for him to acknowledge that the Cycle of Erosion may be flawed. We also argue that the tenets of the cycle were not a unified theory, but a set of regulative principles for governing what is and is not an acceptable geomorphological explanation. We also claim that this set of principles has been replaced by the explicit infusion of physical, chemical, and biological principles into geomorphology over the past 50 years. It is now these principles that regulate what is and is not an acceptable theoretical explanation in geomorphology. One cannot propose and explanation that fundamentally violates these principles - a constraint consistent with Gilbert's approach to geomorphology.

    Bruce Rhoads

    Date: Nov 09 1998
    From: Paul Hesse

    Just an observation on the question of the Davisian cycles;

    Following Richard Marston's message I am prompted to write about my current experience in helping formulate a new syllabus for earth sciences in high schools. One of the responsibilities is obviously to research the area well and to indicate suitable texts or sources (preferably not journal articles which are inaccessible to most school teachers). Consequently I have been flicking through many of the (carbon copy) 'earth system science' lower level university texts. These have their strength in geology and for the most part rely on the uncritical repetition of comfortable geomorphological and pedological models when they stray from geology. Block diagrams of Davis' cycles are common (as are photos of the important weathering process of tree roots breaking down granite boulders!). Clearly the geologists (at least those that write texts) still think Davis is OK.

    Unfortunately many of the geomorphology texts also only give a partial alternative to Davis (so I sympathize with the original message on this topic) but don't necessarily follow through with analysis of how far reaching the influence has been. An example would be the diagram of geomorphic 'equilibria' and time; steady state, dynamic, cyclic (I think I missed one). The last one, cyclic, seems to me to clearly derive from the Davisian model and I can't think of a good reason why it hasn't gone the way of the rest of the model. A paper of Ahnert's a couple of years ago gave a very good alternative, and commonsense, view of the issue.

    It's easy to see why anyone would be confused about 'what the alternative is'. It's also probably a good sign of an active discipline. It would probably be better if texts were more prepared to be upfront about doubts and controversies and take on the shibboleths of traditional geomorphology.

    Paul Hesse
    Dr Paul Hesse
    Lecturer in Physical Geography
    President, Australasian Quaternary Association
    School of Earth Sciences
    Macquarie University, Sydney
    NSW 2109 Australia
    Phone (+61) 02-9850 8384 Fax. (+61) 02-9850 8420

    Date: Nov 10 1998
    From: Gerardo Bocco

    >>What we replaced Davisian geomorph with is not the
    >>dramatically over-simplified denudation chronologies
    >>of landform development, but rather the modern
    >>understanding of landform complexities, threholds,
    >>climate and process change through time, rock (lithologic)
    >>control of landforms, the convergence principles,
    >>complex responses, and many other ideas. Hard to
    >>teach at an intro level, but all least a bit less
    >>misleading to the uninitiated, one would think.

    >>Hope that helps.

    >>Jack Shroder

    1. Denudation chronology is a simplification, as any other model, in this case of landform and landscape development. It has important merits, especially if one wants to do applied research.
    2. Misuse of the Davis approach is something Davis did not want, much alike Wishmeier and the USLE. Probably a good case of misuse is the application of the cyclic model to gully erosion research.
    3. However, if one compares research results of W.M. Davis on the one hand and USGS researchers on the other in the first decades of this century (see below some references), concerning gully initiation in the Great Plains and other areas in the US, may be it is clear that misusing the concept was quite easy.
    4. The ideas of "normality" and "cyclicity" , key to the Davis thinking, are quite difficult to have inspired the concepts of thresholds, complex responses and controls, which have been extremely illuminating. They rather reflect a linear and evolutionary way of thinking.
    1. Brian, K. 1925. The Papago country, Arizona. USGS Water Supply Paper 499, 121-123.
    2. Davis, W.M. 1927. Channels, valleys and intermont detrital plains. 66(1708):272-274.
    3. Fenneman, N.M. 1922. Physiographic provinces and sections in western Oklahoma and adjacent parts of Texas. USGS bull. 730. (see pp. 126-129).
    4. Ireland, H.A et al. 1939. Principles of gully erosion in the Piedmont of South Carolina. Tech. Bull. 633. USDA. Washington.
    5. Johnson, W.D. 1901. The High Plains and their utilization. USGS 21st Annual Report.
    6. Rubey, W.W. 1928. Gullies in the Great Plains formed by sinking of the ground. Am.J.Sci. 215, 417-422.

    Regards, Bocco
    Gerardo Bocco
    Investigador Titular
    Laboratorio de Geoecología
    Depto. Ecología de los Recursos Naturales
    Instituto de Ecología de la UNAM-Campus Morelia.
    Tel. 52 43 200517. Fax: 52 43 200830
    Dirección postal: AP 27-3. Xangari. 58089 Morelia, Mich., México

    Date: Nov 11 1998
    From: Gregory Pope

    It just occured to me that this evolution from Davis parallels the evolution of Astrononomy. Copernicus wanted to portray the universe and planetary motions in perfect circles. A grand model, and simple, and a big improvement on the earth-centered universe, but it didn't quite work. Later astronomers like Brahe tried to fine tune the circles with "epicycles", circles within circles, more complicated, but still trying to fit within the circular motions model. Modern astrophysicists now realize the subtleties of a sometimes complex but ultimately correct process involving gravitational surfaces in space-time. Now translate that to geomorphic surfaces. The complexity befuddles a simple model, but I think the end result, a fantastic landscape based on a myriad of interactions, is just as inspiring to both beginners and experts.

    My apologies to the more astronomically learned.

    Greg Pope (
    Earth & Environmental Studies
    Montclair State University, New Jersey

    Date: Nov 11 1998
    From: JAY PIPER piperjay@VERSAR.COM

    I think we all agree that Davis' concepts have long fallen out of favor. I've been out of school for a while now, so bear with me for a few thoughts on Davis:

    Davis' ideas should be looked at as a product of his time. Is his focus on gradual change and "normality" a vestige of the battle between catastrophism and uniformitarianism? The urge to move away from the biblical flood and "acts of god" as part of geologic history made the geologic consensus shy away from the extremes of magnitude/frequency -- a bias which lingered late in this century, in opposition to J Harlan Bretz' ice dam flood origin for the channeled scablands. Similarly, Darwin said that most evolutionary change probably occurred rapidly in isolated populations --but he did not shout that as "punctuated equilibrium" because he was focused on proving natural selection, and any kind of evolutionary theory, to a skeptical world. So, the emphasis on gradualism was in part an over-reaction in the 19th century effort to substitute 'science' for theological catastrophism.

    Both Davis and Pencke were suggesting predictable landscape changes over time, which could be better described as 'landscape history' than 'landscape evolution'. Evolution connotes optimization, and modern ideas about process/response, fluvial grade, 'dynamic equilibrium', and thresholds are closer to that idea of systems in constant change seeking equilibrium or optimization. Landscape 'history' following a defined cycle is more akin to early misinterpretations of evolution which assumed a 'script' guiding life to higher development. The process/response and dynamic equilibrium models are more like Darwin's picture of a self-adjusting, unscripted system.

    Davis' cycle does seem not too far off as a description of what happens BETWEEN thresholds -- based on looking at a lot of the US and flume fluvial and fan models. Relief and grade decrease over time, most erosion takes place where energy is available -- in areas of high relief or gradient (Davis) and at selected other places like the outside of point bars (not sure if this is Davisian). Lower elevations aggrade and the change of the system becomes limited by sediment transport, so base level becomes a limit to the landscape. The Davisian picture isn't too far off -- until base level is lowered, or uplift occurs, or precipitation/runoff increases, or some other factor increases energy available to the system. The idea of dating a fault scarp by predictable change in its profile over time seems Davisian to me -- dating the abrupt event by the decay of the signature landform.

    Planetary geology has helped to show the extremes that are possible in terrestrial magnitude/frequency relationships for geomorphic events. So we now pay more attention to ice-dam floods, lahars, record precipitation events that trigger multiple debris flows or force changes in channel geometry, etc. These are the high magnitude events that cross thresholds and 'rejuvenate' the landscape cycle, to put it in Davisian terms. The same is accomplished by increased relief from uplift, or access to a lower base level, allowing the release of potential energy stored in the system.

    Where Davis hit furthest from the mark seems to be the idea of the peneplain. It may have been the best he could come up with in the absence of modern tectonics. The question of "which came first - the mountain gap or the river" is more easily answered with understanding of tectonics. Davis couldn't answer that question given the knowledge at the time -- without a picture of plate collision, and drainage systems responding to uplift, the cumbersome peneplains were the best he could muster. Well into this century there were disputes about high-level sediments in Rocky Mountains -- first described as relict basin fills, a kind of single-process Davisian interpretation invoking huge amounts of basin fill "peneplains" -- vs. later characterization as till and outwash, a multi-process interpretation that also considers climate changes in the region.

    Where Davis fell short is

    1. the "big universal theory quest" as John Jansen put it earlier in this thread,
    2. seeking to impose a general 'script' on multiple processes and thresholds within the landscape, and
    3. a fault shared with many of his contemporaries (but not all e.g. Darwin's observations of earthquake and uplift in Chile on Beagle voyage) which was to confuse gradualism with uniformitarianism.
    19th century geology was trying hard to argue for the immense scale of geologic time shown in the rock record, despite opposition from the likes of Lord Kelvin, who said the sun could not be more than a few millions of years old. In the effort to move away from "special creations" and deluvian explanations, geologists were biased away from the catastrophic end of the magnitude/frequency spectrum of geologic events. Davis' ideas seem to follow in large part from that bias.

    Sorry for lack of footnotes, I've enjoyed reading this thread, look forward to more comments.

    Jay Piper

    Date: Nov 12, 1998
    From: Ben Everitt -

    I am enjoying the discussion of Davis started by Paul Larson. Jay Piper scored a hit by suggesting that scientific theories should be understood in their social and political context. Davis lived and wrote at the end of the Victorian era, before quasars, the big bang, chaos theory, the Tunguska event or the October Revolution. The universe was eternal and Newtonian, operating under laws which were (or were about to be) thoroughly understood. Change was understood as "progression"; an evolution toward a better, optimal, or stable state. No one doubted that history was "unfolding as it should."

    It is a rare individual, scientist or otherwise, who has chosen not to sing from the same score as the rest of the chorus. Few have been heard of since. Most, including Charles Darwin, followed as best they could in spite of the facts. I imagine every one of you has had this discussion with editors of peer-reviewed manuscripts. For those interested in exploring this theme, two excellent books are on my list:

    1. Peter Bowler "The non-Darwinian Revolution: reinterpreting an historical myth", 1988, Johns Hopkins University Press.
    2. Arthur Koestler, "The Sleep Walkers: a history of man's changing vision of the universe", 1959, MacMillan. is still worth re-reading.

    Ben Everitt

    Date: Nov 17, 1998
    From: Mario Panizza

    Dear Geomorph-list members,

    Before discussing the subject “Wm Morris Davis”, I would like to precise some concepts concerning “Geomorphology as a science”. It must be said that Geomorphology is an “empirical” science, that is, a science based on observation and experience, experimental research on the various processes and forms of landscape formation, that is, on individual cases chosen among the infinite variety of geomorphological phenomena, all of which are different. For this reason, it is impossible to apply a “systematic” method following the procedures of other natural sciences such as Zoology or Botany. The adoption of methods of this type, that is, the conception of Geomorphology as a systematic science using schematic procedures to explain the genesis and evolution of a physical landscape, can lead, and in certain cases has unfortunately led, to misconceptions and a fossilisation of research, by weighing it down with rigid preconceived models and patterns.

    Should it become necessary to adopt models, they should either refer to specific examples or be obtained from cases that have been directly observed. In this way, they are given no genetic significance but may be used for descriptive and illustrative purposes. Other models may be of theoretical value alone and present an intentionally limited number of variables selected from the infinite number of variables that play a part in the evolution of geomorphological phenomena. For this reason, the sole purpose of such models is to define the role of these variables, considered individually or in connection with some others. However, they should never be taken as rules for indicating specific aspects of morphogenesis. For the same reasons, it is not possible to “classify” either geomorphological processes or the landforms they create. In other words, they cannot be subdivided into specific “classes” like plants, animals, minerals and so on. They may instead be split into groups which present certain genetic analogies, considering a limited number of parameters. A correct and satisfactory procedure for the genetic interpretation of geomorphological phenomena consists in an accurate scientific analysis of the landforms and the numerous causes that have generated them. Simplistic deductions based on exterior appearance alone should be avoided, genetic significance should not be attributed to models that are only descriptive and phenomena that are genetically different should not be grouped together. Geomorphological studies often deal with landforms that have the same appearance but a different genesis, which are examples of geomorphological convergence.

    As regards the Davisian theory, the progress of geomorphological research has demonstrated that most of this theory was based more on imagination than on direct observation of natural phenomena. Illustrative examples, which were first introduced for the genetic description of a specific landform, afterwards assumed a universal meaning, as if they were models of an ideal evolution. This kind of procedure hindered detailed, specific and particular analyses on single landscape forms, thus inhibiting the correct interpretation of natural phenomena and leading to a sort of torpidity in research. On the contrary, scientific investigations proved the importance of landforms and deposits generated in epochs previous to the present and largely due to climatic changes. Other types of research demonstrated that tectonic deformations, with varying intensity in time and space, were among the factors of geomorphological instability. Finally, the growing need for practical applications of Geomorphology, the awareness of man’s influence in changing natural equilibria and his growing role as an agent of geomorphological modelling required more and more detailed, specific and finalised investigations, outside preconstituted general models and averse to the abstractions of fantasy.

    Therefore, modern Geomorphology should be based on observation and experience. In this way the study of processes and landforms and their evolution parameters will allow not only the present dynamics to be identified but also the trend and pattern of future evolution to be predicted. This methodological approach is by the way indispensable for a correct interpretation and solution of the problems concerning the practical applications of Geomorphology.

    Mario Panizza
    Vice-President of the International Association of Geomorphologists
    Dipartimento di Scienze della Terra
    Università di Modena, Italia

    Date: Nov 18, 1998
    From: JAY PIPER [piperjay@VERSAR.COM]

    Mario Panizza did a good job of pointing out the misguided effort to use Davis’ theory as a systematic theory of geomorphology.

    Davis came up with a pretty good conceptual model for his home ground -- the Appalachian Plateau. His concepts, summarized by the infamous block diagrams, worked well to explain this region which can be generalized as follows:

    His model probably gained credibility as reports came in from westward exploring expeditions, since parts of the Colorado Plateau can be also be at least superficially described as above.

    His model foundered when applied to the Valley-and-Ridge Appalachian folded terrain east of the plateaus. Trying to correlate ridgetops with past erosional surfaces (‘peneplains’) resulted in the multiple peneplains listed in turn-of-the century literature. The relationship of drainage and folded ridges was better explained by fluvial adjustments during deformation (I don’t remember the reference for a paper comparing the Zagros mountains of Iran and the older Appalachians, examining drainage development) than by Davis’ invocation of peneplains and superimposed drainage.

    In talking with other geology students, we were puzzled by two things as we studied geomorphology:

    1. where was “the rest of” Davis’ theory-all descriptions seemed vague or incomplete
    2. why was Davis often painted as somewhere between a fool and a villain?

    In retrospect, the theory seemed incomplete because there wasn’t much more to it than the notorious landscape cycle block diagrams. It was a reasonable conceptual model for some combinations of stratigraphy, tectonics and climate history-and it was overextended by application to deformed or glaciated terrain, complex process or climate histories, etc.

    Davis was not painted kindly because he forget the field basis of geology --apparently he fell in to the trap of arguing about the facts not fitting his theory, instead of remembering that any theory needs to fit the facts seen in the field. In other words, “he tried to see it because he believed it” to paraphrase the famous saying. Given the struggle to by Darwin, Gilbert, Lyell, Hutton and many other ‘heroes’ to replace catastrophism and aristotilian speculation with an empirical, defensible body of earth science, it’s no wonder there is little professional sympathy for Davis.

    On the other hand, he worked in relative vacuum with little understanding of plate tectonics, dating methods, fluvial geomorphology and climatology. His block diagrams of the landscape cycle were not as great a leap of insight as Darwins’s model of coral reef development, but like Darwin’s reef model they should have remained in the same descriptive, limited-application arena, rather than being stretched into a unifying theory. I don’t know enough about Davis’ biography - did he burn professional bridges in applying his theory or try to squelch alternative ideas? He was after all a contemporary of Cope and Marsh, who practically had an armed feud ongoing in American paleontology.

    Jay Piper -

    More opinions

    Back to Theoretical geomorphology

    Back to The Virtual Geomorphology

    Edited by The Association of Polish Geomorphologists
    Last updated: 1.10.2002