What makes a good discussion topic?
Posted by Leon Follmer 22 Aug 2005 22:59:18 PP DISCUSSION What makes a good discussion topic? Outmoded standards Over the years I have become concerned about a growing disregard for standards that are outmoded, or not suitable for modern research. I am sure everybody is familiar with some old standards or conventions that prompt a feeling “that it should be changed”. Because standards The Paleopedology Commission is in a position to be able to address some of the issues. In this light I submit an example for examination, a review, which discusses a suite of problems important to paleopedologists. The review is my comments on a manuscript submitted for publication in a leading Quaternary journal. The paper failed in my opinion to reach its goal because old standards and conventions made it impossible. More specifically old standards force authors to use or to do things that they know or should know are not right or best choices. When conceptual tools are limited then the results will be limited. Are the authors responsible? In part, they are, but the solutions may seem to be out of their control. The paper discusses the familiar Holocene “loess-paleosol sequence” of the Chinese Loess Plateau. I understand better now why the term is hyphenated – some people don’t really know the difference. Because so many parts of the manuscript pointed to language and concept problems I thought it I would be best to go into basic paradigm issues to find common grounds. At this point I realized that this is a subject and in the form worthy of a discussion subject for the PP Commission Discussion web page. I excerpted the commentary appropriate for the Discussion page. ………………………………………………………………………………………………… General assessment: The research goal of this paper is a significant issue for China and has application to loess-paleosol work in other parts of the world. Most of the findings and arguments are appropriate but are seriously flawed by basic language problems and incomplete treatment of some issues. At this point the manuscript is unacceptable but it can be salvaged. Several issues concerning content stand out, which I will address here in a form of a prologue. 1. Loess - paleosol confusion in China Some of the best work on loess-paleosol sequences in the world has been done in China. But like most earth science work, the initial framework of models and classification schemes at some point become unsuitable or incapable of handling new information and discoveries. What is so clear in this manuscript is the inadequacies of the traditional Chinese method for naming and classifying soils and loesses, and in particularly their subdivisions. They are different in their nature and origin. They need to be treated on their own separate terms and not be considered equivalents. To clarify this point the Chinese style describes a loess-paleosol sequence as an alternating series of paleosols [=soils] and loess units. Loess is the parent material for these soils so the meaning and relationships among the terms and derivatives are rendered ambiguous. Those of us who have seen the “sequence” quickly realized that the “paleosol” is in most cases a Bt horizon of a paleosol developed in loess. The “loess” is either a C horizon of a loess profile or is the upper solum of a soil profile with confounded A and E horizon characteristics, which is often overprinted with B horizon features from the overlying paleosol. The important relationship that I observed on the Loess Plateau is the commonplace of continuous pedologic features from one horizon to another in typical profiles. Related to this is the accretionary nature of the paleosols where the upper solum [A+E horizons] are stretched upwards [overthickened], and continue up to an overlying B horizon. This terminology problem stems from a language problem described by some as nonparallel construction. This is the consequence of two items [nouns] that are not in the same category nor are logically comparable. For an analogy, think of loess as a vegetable and soil as beans. It is confusing to compare and distinguish beans from vegetables. However, this analogy does not go far enough. Loess and paleosol are more different than this. As you know loess is a material in which soils can form. And I think that you should recognize that these concepts need to be kept separate for scientific reasons. In former times when the loess-paleosol concept was treated as a simple generality, no problems arose but now when you choose to study specific features of each you must use appropriate language to make meaning and relationships clear. In your work you have reached the point where distinguishing the geologic attributes of loess and the pedologic features of paleosols is critically important. For modern work following modern standards, it is necessary to recognize the diagnostic parts of a soil [soil profile], which are the A, E, B, and C horizons, so that you can distinguish them from the loess units and subdivisions of the loess units. In other words, the parent material factor in the five-factor soil formation model should be treated rigorously. Because pedologic features differ from horizon to horizon, it is not possible to make much sense out of your discussion comparing “soils” as if they are internally uniform. Long ago I could see that the time will come when rigorous and separate criteria for loess and paleosols in the study of the Chinese loess record will need to be implemented. Continuing old generalities is no longer justifiable. In most of the scientific literature today this technicality of discriminating parent material and pedologic characteristics is already being practiced. To make your paper modern you need to identify the soil horizons as well as to identify the pedologic features in your thin sections.
The L and S stratigraphic terminology scheme was logical in the beginning when there was only one hierarchial rank, which by default placed loess and soil in the same class. Then modifications were made that added subscripts for subdividing the principal units. The expanded scheme was still workable, but the subscripts became a mark of “sophistication” and uncertainty. Then when multiple letter and number designations and superscripts were added it marked the end of the line for “nonspecialists” to understand or even care about what they mean. Do you want readers to understand, or readily grasp? Then strip away the technical appearance of the symbolism and do the following: use the basic L and S numbered structure and then announce that for the purposes of discussion in this paper you will arbitrarily number the S and L subdivisions. Actually, you can use informal names or propose formal names for subdivisions that have qualities of “key” beds. These are “layers” the average man can see and trace from site to site. Also, don’t use soil taxonomy names for stratigraphically defined paleosols [geosols]. It should be obvious that this causes confusion. Soil types within a geosol, a stratigraphic unit, need to be free to vary along a traverse, a catena, or across the Loess Plateau. As a rule geosols encompass many kinds of paleosol types. The types in a geosol need to be recognized for what they are rather than insinuating what kind of soil they are by labeling a stratigraphic unit with a soil type name. In a material-stratigraphic scheme Geosols serve as place holders. For full treatment of this issue, a dualism, the parent material needs to be identified or interpreted and assigned to an appropriate material stratigraphic unit, for geological reasons. In other words paleosols in the geosol sense serve two roles, one in soil stratigraphic classification and one in material stratigraphic classification. Also, don’t forget that this distinction is independent of what kind of soils they are, which is another issue. In stratigraphic work, pedologic characteristics are used to distinguish the part of the geologic material with pedologic features from the part without them. In some cases a soil profile and a material unit correspond one to one, but this does make them equivalent within a classification scheme. This relationship is often assumed but is usually wrong. Only in special circumstances does the lower boundary of a B horizon correspond with a lithological discontinuity [=geological boundary], or any feature that one could use to argue that the boundary has both attributes. Thus, soil profiles are superposed onto geologic materials or have overlapping relationships. Being a soil does not eliminate the parent material attribute on which it is classified in a material classification scheme. In this sense soil is not a material but an entity formed in a material. Don’t forget, parent [=geologic] material is the fourth factor in the famous soil-formation equation, and each factor needs to be independently assessed for scientific reasons. 3. Static age of paleosol fallacy There is no word for this but for people who generalize that a soil is a “unit”, also assume by default that all parts of the soil are the same age and have the same character. In specific terms, this generalization is never true. Only parts of a soil have the same age or form contemporaneously, however many processes do overlap in time. The main features in a mature soil have long histories and are full of interruptions. After all, climate [change] is a principal, if not the most important, soil-forming factor. In spite of the fact that many soils look simple they in fact have complex histories. In the simple sense soils grow downward as soil horizons thicken. Lower boundaries of horizons are analogous in some ways to the roll front concept in geochemistry. As a change of conditions encroaches and passes over older soil features, it changes some of them to younger features, and commonly leaves behind relics. In general in an active soil, the lower B boundary is younger than the upper B boundary at any point in time, and so on with the other horizon boundaries. In your case you have the added complication of soils that grew upwards as well as downward. In this scenario the upper B boundary grows upwards into the former E or A horizon as the upper solum thickens. Whether it is an E or A is a significant observation; it is an indicator of the paleo-environment. There are many articles and books about these subjects. These concepts are important and necessary for your work. Comments: I do believe that the interpretations of the stratigraphy and pedogenic history in this manuscript are on the right track. Also, placing emphasis on an integrated concept of loess additions and pedogenic modifications is modern thinking. This could be a seminal paper about the loess-paleosol sequences of China, but it falls short of this goal. A main problem is confusion surrounding the discussions about soil and stratigraphic features and their equivalencies. The conceptual elements of soils as they are treated in this paper are archaic. This paper perpetuates an old problem that is dependent upon an inadequate model of stratigraphic classification of loess and on an unrealistic model of “soils”, which treats them as uniform static entities. The time has come when this needs to be changed so that the details of the Holocene loess-paleosol sequence can be explained in a more meaningful way. L R Follmer 19 August 2005 I
|