Sir,
Geoffrey Boulton answers my letter on the contributions of J.G. Goodchild to glacial sedimentology by two thrusts. First, he avers that Goodchild’s under-melt interpretation is largely incorrect. Secondly, he alleges that some glacial geologists ignore actual physical processes. I find his first argument to be unsubstantiated and his second unwarrantable.
If Boulton is to demonstrate that Goodchild’s hypothesis is wrong, he is obliged to show, by means of evidence, that it is contradicted by either observation or sound theoretical principles. Alternatively, he could show that another hypothesis better explains the original and any subsequent observations on the glacial deposits of the Vale of Eden. He does neither in his reply and we are asked to discard an important hypothesis on the basis of an unsupported belief. In a similar vein, Boulton’s dismissal of the under-melt hypothesis, first proposed by Goodchild, is on the basis of opinion not evidence. Let him expound his theoretical reasons for discarding this hypothesis and explain by other means the observations used to support it. There is a world of difference between alleging that something can be done and actually doing it.
I cannot imagine any glacial geologist knowingly ignoring actual physical processes. The actualistic works of Boulton, Lawson, Powell, and others are widely cited in the glacial sedimentology literature. But, when the evidence speaks against known processes, land-form and sediment interpretation requires imaginative inference. Even then, Goodchild did not “fall back on imagination alone”. His interpretation of sand and gravel as a product of water washing was founded on experience. His method was both experiential and inferential, involving the interpretation of well-documented observation. I find it odd that this method could be dismissed as leading to a dead end.
Boulton employs the technique of interpreting my meaning to make it appear absurd. Such cleverness may win debating points, but it does nothing to promote understanding. Evidence from the direct observation of process is desirable not because, as Boulton implies I suggest, it is an optional extra, but because it is a luxury not always available to geologists. As examples, the theory of continental drift, the flood hypothesis for the Channeled Scabland, and much of our knowledge of the behaviour of deep-sea turbidity currents all originated without the luxury of direct-process observations.
I am under no illusion that reconstructions of the past are real; we are concerned here with validity not truth. This humbling thought applies to any explanation of ancient sediment and land forms, whether we appeal to known or deduced processes. But, since Boulton sees fit to question my sense of reality, I feel free to examine his. He claims that, whereas his concept of subglacial melt-out till (Reference BoultonBoulton, 1970) is based on observation, others base theirs on inference alone. Despite this assertion, I find no direct observations in Reference BoultonBoulton (1970) on the process he defined as subglacial melt out. On the other hand, he described 2.4 m of till he believed to have been deposited subglacially by the melting of stagnant ice beneath Nordenski�ldbreen. This till is clearly of melt-out origin according to the definition used by those whose thinking he finds flawed:
Melt-out till — till formed by the melting of debrisrich ice that is neither sliding nor deforming internally in the zone of formation (Reference ShawShaw, 1982, p. 1549).
Reference Boulton and GoldthwaitBoulton (1971) caused confusion by referring to this subglacial till as a type of lodgement till. It appears to me more logical to classify it as subglacial melt-out till, with the consequence thatBoulton’s argument against the under-melt principle is also an argument against his own observations. I believe he is wrong on both counts. I find no fault with the conclusions he drew using Goodchild’s tradition of inferring process from sedimentary characteristics. The thermal and dynamic regime of Nordenskiöldbreen over the 200 years or so of till deposition cannot possibly be described from a synoptic view of the ice/bed material interface near the present glacier margin. Thus, Boulton’s claim to objective reality is spurious; like Goodchild's, his conclusions depend on inference.
I find J.G. Goodchild’s influence to be of lasting significance and his under-melt hypothesis remains unchallenged. No evidence has been presented to the contrary. Despite his claim to a broad view of sedimentology, I believe that Geoffrey Boulton’s observations are so restricted in time, geographical extent, and environmental context it is hardly surprising that he recognizes rather limited sedimentological associations for melt-out till.
