A guest post from Dave Foster of Naval Air Systems Command, China Lake. The views expressed in this post are those of the author and do not reflect the official policy or position of the Department of Defense or the U.S. Government.
Recently, we’ve learned from the CNO that our combatant commanders think they need 450 ships to cover all the nation’s operational requirements. Of course the rub is that the Navy’s annual 30-year shipbuilding plan is metronomically panned every year as budgetary fantasy. The Navy’s FY2014 plan is a 306-ship plan if fully executed, which it won’t be. Even if the Navy devises a much more liberal ship counting approach than what they’ve used to come up with the upwardly revised current inventory of 289, there is simply a snowball’s chance of seeing a 450-ship U.S. Navy in the foreseeable future. Pointing a finger at this troublesome possible future is Representative Randy Forbes (R-Virginia), who claims that today’s Navy can only meet 43 percent of COCOM requirements as compared with 90 percent back in 2007. (The Navy had 282 active ships in 2007 and its FY2006-07 plan was a 313-ship plan). What scant percentage of tomorrow’s COCOM requirements could be met by only two-thirds of the necessary Navy is a mystery, but the implications of this new news, if taken at face value, are grim.
For its part, the U.S. Air Force, presumably responding in similar fashion to COCOM requirements, has maintained a procurement goal of 1,763 F-35A steadily since the late 1990s. F-35A is currently supporting zero percent of COCOM requirements, but presumably all 1,763 of the USAF’s F-35A’s figure into future planning scenarios. But whatever course of action (COA) analyses fed this new need for a 450-ship navy have not, apparently, driven a change in demand for F-35A. I’m not picking on the USAF arbitrarily; both the Navy’s and the Marines’ F-35 variant buy numbers have changed since late 1990s JSF program predictions.
Hmmm. What is going on here? Were the OPLANs changed? Has DoD’s Global Force Management tool been hacked? Did some COA studies at Newport, NDU, and elsewhere come off the rails? For over a decade, the Navy’s shipbuilding plans have shown steadily declining end strength in the out years but now we have this sudden 50 percent spike in requirements. Meanwhile, the USAF’s F-35A buy plan has not wavered thru a decade and a half of strategic and economic turmoil.
It is possible that the overarching force structure requirements for 450 ships and 1,763 F-35A are loosely coupled or even unrelated in our strategic context? Neither the steady decline followed by the recent sudden spike of naval requirements nor the unwavering procurement plan of F-35A over a decade and a half make sense either independently or in conjunction. So, in some way, resourcing estimates are, and perhaps have long been, more generally untethered to any realizable strategic and budgetary reality, and the plans-reality mismatch is as alive as ever.
First, let’s look at the big picture. The full suite of goals in our national strategy and military posture looking into the future are never explicitly defined and substantiated in a single overarching document or plan. Rather, these are expressed in limited and inexact form across a variety of formal strategic documents (some classified and thus unseen), formal pronouncements, informal suggestions, and innumerable inferences and ongoing chatter. Taking a slice from the ranks of formal strategic documents, we can see a shift in rhetorical emphasis over the past decade or so as expressed in the always controversial Quadrennial Defense Reviews (QDR). This is, admittedly, a somewhat tenuous means of survey but it may suffice to establish a broad point about the apparent mutability and evolution of national strategic goals over even a relatively short interval and a disconnect between strategy and resource planning.
A brief review of the QDR series is in order. The 1997 QDR was very numerical and included explicit force structure numbers and types of units for each of the services. The strategic emphasis in 1997 was on the requirements for full spectrum of conflict capability as necessity for “a global power with global interests to protect.” In this was the high water mark of the American unipolar cognizance. The 2001 QDR focused on the capability to conduct major combat in two regional theaters with an emphasis on decisive victory in one. The 2006 QDR viewed stability operations and combatting and mitigating terrorism and WMD as our nation’s essential military tasks. The 2010 QDR was a bit more gauzy in emphasizing the need to prevail across a complex range of time, space, and adversary while the 2014 QDR laments our capability restrictions linked to financial limitations and distressingly predicts that the U.S. will have trouble responding to more than one “major” contingency at a time.
One could thus say that the United States government has recently experienced a philosophical migration from confident, if not strictly dominant, global power to impoverished and tentative global actor among many. One might observe that this has occurred in the final eight percent of the time interval since America declared her global presence and interests through the Monroe Doctrine. One could further say that this has nevertheless implied demand for a certain generalist major power military capability. And in turn, this requirement for general capability implies certain things about force structure, its global allocation, and the nation’s optimal development and procurement strategy.
One could say. But we know that DoD, preeminent of all the major militaries in the world, is a systematizing and rationalizing organization, and generalist strategic philosophies are not distilled into general plans for inexact numbers and types of people, things, and skill-sets in its arsenal. Rather, DoD’s planning has been and will be very explicit, and the result of continuous refreshing via the detailed examination of alternative courses of action and an attempt to match perceived threat capabilities to desired U.S. capabilities. But just as knowing is not doing, planning is not knowing.
The planning-knowing gap has been expressed in different terms in the familiar innovation literature. Moore’s chasm and Gartner’s trough of disillusionment are two ways to express it. Moore’s pitfall is most apparent in the swirl of R&D to idea design maturation where the practical usefulness and realizableness of ideas is ambiguous and abstract. Gartner’s trough is always looming as the hangover consequences for seemingly bold and interesting ideas and has been a particular feature of long-running wars of ambiguous rationale, as McNamara suggested in In Retrospect, and in long-running weapons development programs of ambiguous cost and capability.
DoD’s planning-knowing gap—and this is directed at all forms of planning—occurs in going from highly specific synthesis and analysis, where physics and even the plausibly quantifiable data of Likert surveys for trade studies and war games are fairly tangible, to rolling up the results of dozens of these subsidiary plans into plausible ideas about knowing what can be operationally accomplished. The gap gets wider in proceeding from the plausible to what may be strategically realistic.
A recent post on this blog pointed out that one’s military resource choices are ideally matched to one’s most plausible expectations for war. But how are these choices made? Comparative weaponeering of blue and red measure/counter-measure capabilities is one of the more science-y elements of planning for where the rubber meets the road in combat. Political science and evolving international norms, such as they are, comprise many of the more nebulous aspects of planning. Above all, strategic means-ends planning is a wicked problem of interwoven synthesis and analysis where not much is really known about the ultimate goal/end-state nor are there much more than unevenly distributed bubbles of data, information, and technique within a flowing opaque matrix of imprecision and uncertainty. So, seeking traction, planners reach for tools and theories about the value of the tools. But the tools are better for the local and specific and are problematic as the level of abstraction rises.
Here’s an example from one tentacle of the complicated resource planning task.
Start with munitions lethality estimates, a core of operational weaponeering. Lethality is an unfortunate simplification of operational capability but, face it, when the president asks where the carriers or the SSGNs are or when an amphibious ready group is vectored off shore from some trouble spot, much of what is practically thought of by the president on down in these cases (non-combatant evacuations excepted) is strike capability.
The DoD plans for strike by starting with estimations and observed data gather through testing of the explosive performance of warhead configuration X. Knowledge of chemical and material structural properties along with test data help to determine nominal distributions of fragment size, fragment velocity profile over distance, and the pressure and heat effects of the blast. An explosion is neither an ellipse nor the multi-pointed star seen in cartoons—but at the same time, it is in some ways a little of both. Additional information on the expected accuracy of the warhead delivery system and the precision of the system’s physical orientation at impact are important. Once evaluated, these estimated and observed data can become information, which allows for some degree of lethality performance estimates of warhead X against various targets of interest.
Even at this stage of evaluation, there have been many necessary judgments and assumptions to fill in the lack of measured parameters taken under a range of operationally relevant conditions. Thus, and obviously, it will never be known where 100 percent of the fragments were across the time interval from detonation to whatever becomes the new equilibrium settling out post-explosion. And there will only be pressure and heat data samples from instrumented intervals at selected clock positions and distances surrounding the detonation point.
From here the evaluation becomes largely statistical and samples of data are generalized into estimates of lethality against targets of interest based on the frag/pressure/heat sample data and various orientations of warhead versus target or warhead penetrating target. The final weaponeering models become, like all models, abstractions of general cases of certain parameter interactions. Morris Driels of the Naval Postgraduate School has an excellent text describing this weaponeering process.
Ultimately, the operating forces can weaponeer from the result of the art of choosing the best way to interpret and compile a vast amount of partly empirical, partly statistical, and partly subjective information and create useful but imperfect models. The operating forces can estimate the efficacy of warhead X delivered to proximity with threat target Y by (often enough) “precision” guided munition Z via platform W which has survivability A verses threat air defense system B. From here, the planning activities of the COCOMs, Joint Staff, service staffs, war colleges, FFRDCs, and various other activities can make more abstract estimates based in part on these weaponeering estimates and can begin to develop force structure capability plans for whatever COAs are to be investigated and possibly planned. At the level of COA contingency planning, during training, and in actual war, weapon X usually only has a general lethality against target Y even with the use of tactics, techniques, and procedures that are developed to optimize the relevant employment parameters for the relevant operational conditions. In short, the exact lethal effectiveness of X versus Y will not be known and the efficacy of the overall campaign plan becomes in large measure probabilistic.
Some bubbles of information in the flowing matrix of uncertainty and imprecision are based on measured behaviors of physical systems in operating in some sector of designed operating envelope. Some are based on quantified social science. Amid the fog of planning and war, much of what is “known” is really just derivative of relatively distant and partial data which has become general and is indeed fairly abstract. In other words, the effectiveness of the carrier group or missile sub is by the high level planning/gaming stage a general match up against a target set developed for desired operational and strategic purposes. To check the likelihood of this assertion, ask the following question: if weaponeering was sufficiently robust to be deterministic in effectiveness estimations, then that last step of the kill chain, Assess, would be unnecessary.
And so this goes on, for all the warfighting systems and units with different details for other diverse effectiveness metrics like reliability, maintainability, survivability, proficiency, sustainability, deployability, training, manpower management, etc.. All evolve over time—and towards obsolescence—with periodic evaluation through the lenses of the physical sciences, social sciences, operational art, and strategic philosophy.
Yet from these well-considered strategic generalities about unknown and unknowable future end states, we get explicit numerical resource requirements like 450 ships and 1,763 F-35As. We get such implausible exactness for sensible rationalist reasons such as budgeting and manufacturing planning and in the resource allocations for plausible or expected blue-red match-ups. But history suggests that we should always be skeptical of any digging-in-of-the-heels on specific numbers because these numbers are really just one of any number of plausible answers to the stacks of assumptions, interrelationships, sensitivities in the questions. Indeed, budgets are a constraint in the real world and should inform the plausible strategic futures that DoD plans for something that neither the 450 ship or 1,763 F-35A, to name two examples of what may be a greater set of implausibly specific resource futures.
Perhaps America’s belief in the tools is an aspect of what Isaiah Wilson III finds in the nation’s preference for “happy endings and no loose ends,” in its tendency to invert the order of cause-and-effect in the relationship between power and technology, and in its “distaste for tragedy [that] has led US strategists and policymakers to mistake mere force for power.”
This is not an argument against planning, but a critique of the precision and usefulness of what we think or claim we know and what is knowable. To plan and develop tools for planning is sensible. But knowing what to do, let alone doing it, does not necessarily follow from the planning effort. I suggest that the military, the congress, and the public should be more circumspect in their belief in the tools, because the historical record of strategy-resource planning in providing a foundation for knowing what to do is unimpressive.