I was pleased to have Bryan Clark and Dan Patt join me on the Acquisition Talk podcast to discuss a range of important subjects including mosaic warfare, joint all-domain command and control, the digital century series, systems architecture, and more. Bryan Clark is the Director of the Hudson Institute’s new Center for Defense Concepts and Technologies, where he is joined by a former deputy director at DARPA, Dan Patt. The two make a formidable team, and their center will focus on the application of emerging technologies to military concepts of operation.
In the episode, Dan and Bryan argue that the DoD’s reliance of monolithic platforms — a relic of the Cold War era — makes it increasingly fragile to defeat against peer adversaries. Major systems today are expected to perform numerous missions, requiring them to self-contain sensors, command and control, combat systems, and so forth. This not only increases unit costs and decreases force structure, it limits the number of different ways force packages can be composed. US commanders are thus limited in their options for effecting a result, making them much more predictable and subject to counter-measures.
The alternative is to decompose monolithic platforms into a wide array of smaller systems. While each system itself has fewer capabilities and is less survivable, there will be far more of them. Their lower cost allows them to be attritable. The benefit is magnified by increases to competition and economies of scale. Importantly, commanders will have far more options to decompose and re-compose the force structure.
An important element in the mosaic warfare concept is joint all-domain command and control, or JADC2. This can be described as the network that connects the relevant nodes of the disaggregated force structure, such as between sensor and shooter, and is often called the military “internet-of-things.” With a rise in the need for interfaces, we discuss a path forward to create ad-hoc interoperability between unique system requirements called STITCHES. This by-passes many of the rigidities faced in the pursuit of an agreement on global standards common in today’s modular open systems architectures.
Podcast annotations
Here’s Bryan on the motivation for a paradigm shift in the DoD from attrition-centric warfare to decision-centric warfare:
I would argue that the Chinese are not offering a force designed pursue attrition-centric warfare. They moved up the value chain themselves and are pursuing more they would call system destruction warfare, where they will go after the system of systems that the US developed with its allies, identified it’s weak points: logistics, command and control, information, and communications, and then use that as a way to defeat us by forcing us to accept fait accompli, or accept an outcome that is less advantageous. It’s about affecting our decision making…
We need a force that’s design to impose that kind of decision challenge on the Chinese. This force set translates from the complication of individual multi-mission platforms towards a complexity networked force that’s more recomposable, is the transition that allows us to regain a decision advantage.
What mosaic warfare brings to the fore is logistics. As DARPA’s Jim Galambos said, “now I’ve got more things spread apart, how do I get them there and how do I sustain them?” But logistics isn’t just about moving and sustaining stuff, but moving information through a command and control system.
I think it’s natural to envision the complex network of disaggregated systems being controlled through a single command and control system. Perhaps something like seen in Ender’s Game. The Air Force’s JADC2 concept — currently being implemented through the Advanced Battle Management System — seems to be that single place where all information is brought together. The very idea that there would be multiple and overlapping C2 systems seems counter to the intention of “joint all-domain.”
One of my thoughts is that mosaic warfare should be a fractal concept, or that the disaggregation of systems should also apply at the highest level of analysis. While it seems inefficient, creating one C2 system to rule them all creates a new existential risk: that the enemy accesses it through hacking or otherwise. The focus on just one system also means most of the tradespace goes unexplored, leading to inferior developments.
Here’s Dan on interservice rivalry in C2 systems:
I think it’s very unlikely that the United States can sustain an asymmetric advantage from having a global solution to anything. I’m an extreme skeptic on one command and control system to rule them all. I think the cultural advantages of the US are likely to be things like a commander’s initiative and a commander’s autonomy. In general, the cultural differences between services. Those are the kinds of things where the fact that we’re different, the fact that we’re diverse, those are the kinds of cultural differences the US should seek to amplify.
I fear that the DoD suppresses those cultural advantages the US has in the private economy. There is constant pressure to choose the single-best system very early in the process rather than try out new things. And the fear is that the Chinese, which used to have a reputation for copying American products, are now in some cases on the frontier and even developing new business models.
Another effect of systems disaggregation is that it increases the number and effort required of interfaces. Smaller platforms will swap in-and-out different mission packages of subsystems and components. Each of them needs to interoperate and communicate with the C2 system. The current thinking tends to focus on modular open systems architectures, where global standards are creating for defining the interactions, such as between a sensor and a shooter.
Yet those global standards are very hard to develop and get agreement on, and just as hard to update. Here’s Dan:
Just look at history for a minute. Link 16. How long has that been around? Around 40 years. It is still not rolled out across the entire Air Force inventory. And there’s different implementations of it. The history suggests that, even within one service and one application, global standards are really hard to pull off.
…. You’re hinting at, there must be a better way, and indeed there is… Maybe I shouldn’t think about interoperability as trying to achieve a universal language. I shouldn’t try to define Esperanto, and force everybody to talk Esperanto. Maybe I should do on-demand translation — the Google translate equivalent… It might seem less elegant, but it’s a lot more practical. And [DARPA’s project] STITCHES is the equivalent of that, but for systems.
Listen to the whole thing, there’s a lot more on this concept of ad hoc interoperability, which actually made it into the FY2021 National Defense Authorization Act, which has been big on pushing for modular open systems, and this year directed the use of STITCHES “if appropriate and available.”
Here was a good piece from Bryan on the digital century series — (which he discussed before DARPA’s simulated dogfight saw an AI system defeat a human pilot five to zero):
We think focusing on the manned fighter is fundamentally flawed. The whole purpose of these new technologies is to open up the space of, what is even the role of a person in the air combat space? Is the person a command and control node rather than an operator of an airplane? That’s probably where we’re going. So we think the digital century series should instead focus on the technologies that they really want to address: autonomy, AI, unmanned systems.
Thanks Dan and Bryan!
I’d like to thank Dan Patt and Bryan Clark for joining me on the Acquisition Talk podcast. Be sure to read their excellent CSBA paper on Mosaic Warfare, as well as their various Breaking Defense articles including on JADC2 and the Digital Century Series, They have an excellent OTH article on Decision Maneuver. Bryan is featured on this article on the Navy’s next-gen fighter. I’d also like to thank David Gerson for pointing me in their direction, and for sending me these two briefings on STITCHES (which I’ve uploaded here and here).
Leave a Reply