To the Editor

During the course of the twentieth century, roof falls killed more than 35,000 U.S. underground coal miners. A key episode in the struggle to reduce that terrible toll was the mid-century transition from timber supports to steel roof bolts. Yet despite its rapid adoption by the mining industry, roof bolting failed to have any noticeable effect on fatality rates for nearly two decades (see fig. 1). Why did such a promising and widely celebrated safety innovation take so long to start saving lives?

Mark Aldrich’s new article “Engineers Attack the ‘No. One Killer’ in Coal Mining: The Bureau of Mines and the Promotion of Roof Bolting, 1947–1969” (T&C 57, no. 1 [January 2016]: 80–118) explains the lag this way: “… companies needed to learn how to employ the new technology under varying mining conditions” (p. 82). Aldrich places the U.S. Bureau of Mines at the center of this process of technological evolution and after 1965 “miners’ risks from the ‘No. 1 Killer’ began a long decline.” The contribution of mine safety legislation is relegated to footnotes.

This narrative is at odds with the historical record. The most obvious contradiction is that figure 1 does not show “a long decline” in roof fall fatality rates, but rather a precipitous drop that began in the late 1960s and ended by the late 1970s. Aldrich’s narrative is also at odds with the Bureau’s own contemporaneous presentation of the issue. In fact, it is clear that the Farmington Mine disaster and the subsequent 1969 Coal Mine Safety Act were the proximate causes of the sudden decline in roof fall fatalities. The Bureau of Mines was indeed central to the process, but its role was to forcibly change the safety culture of the mining industry using roof control concepts that long predated roof bolting.¹

Soon after it was founded in 1910, the Bureau concluded that “a condition responsible for many fatalities from falls of roof is the absence of any policy on the part of management with respect to systematic methods of roof inspection and support.” When mine management left decisions about when, or even whether, to install roof support up to individual miners, roof that appeared “good” was often left unsupported. Encouraging mine managers to prepare, promulgate, and enforce systematic roof support plans became the centerpiece of the Bureau’s roof control efforts.

Fig 1.

Fatality rates from roof falls in U.S. underground bituminous coal mines. (Source: Mark, “The Introduction of Roof Bolting to U.S. Underground Coal Mines.”)

Aldrich derides this approach, saying that “Bureau engineers did not know how to study roof falls. They published numbers of largely empirical studies of company practices, but nothing came of them” (p. 86). In his view, roof falls were a problem in search of a technological solution.

This view, however, is based on a misconception of how roof support systems actually work. Stated as simply as possible, roof supports can only prevent roof falls if enough of them are installed. The spacing between the supports is actually much more important than the type of support. If the gaps between the supports are too large, then slabs of rock can fall between them. And no support can protect miners if it has not been installed at all. Bureau studies in the early 1950s found that 50 percent of all roof fall fatalities occurred in the unsupported “danger zone” directly in front of the freshly mined coal face, and 40 percent still did in 1967. Roof bolts generated excitement at the Bureau of Mines because they could be installed right up the mining face without impeding equipment, not because they were inherently better supports or because they were more amenable to engineering design.²

Roof bolts were more expensive than timber posts, however. Each mining company’s understandable tendency, therefore, was to adjust its expenditure on roof support to achieve an acceptable level of roof fall risk. Aldrich cites Bureau records describing “a parade” of mine operators who increased the spacing between bolts to save money, or who...