Jonathan Rauch: “Political Realism”

Political Realism: How Hacks, Machines, Big Money, and Back-Room Deals Can Strengthen American Democracy
By Jonathan Rauch


I read this after AnomalyUK mentioned it on Twitter. I was skeptical because of the two-part title, what Neal Stephenson called “one of those American books where once you’ve heard the title you don’t even need to read it”. But it was short and free (probably why AnomalyUK read it).

And it does start badly, writing in a silly tone as if its audience were idiots (“political home truths”). It’s published by the Brookings Institution. Rauch also redefines “political realism”, seemingly unaware that the term is already used to describe the views of Bernard Williams and Raymond Geuss.

Rauch’s argument is that movements in America to make politics less “corrupt” have been a bad thing on balance, because they reduce the effectiveness of the government. He argues that the definition of political corruption has expanded: “what began as a necessary rebalancing acquired its own momentum and ideology and created an imbalance of its own, pushing relentlessly to extirpate corruption and defining corruption ever more broadly… And today it defines corruption very broadly indeed.”

Corruption is defined in modern parlance as anything which is not democratic. (Rauch doesn’t spell this out.) Money in politics is corrupt because some people have more of it than others, so not everyone’s “voice” is counted equally. Secrecy in politics is corrupt because it prevents voters from being informed about what’s going on. (I prefer Moldbug’s definition: “Corruption is any human action that is not what it appears to be.” More on that later.)


Rauch advocates “machine politics” (basically, party politics with less democracy) as being better at governing than democratic, transparent politics where ordinary people have more power and machine politicians have less. Political machines act like companies; without them there would just be individuals. Parties have longevity, so they have long-term incentives. “They have to win not just once but again and again, which gives them quite a bit of accountability to their stakeholders”. “By trading on the present value of future political returns, they can conduct transactions across spans of time–something very hard, frequently impossible, for activists and freelancers to do.” This is a good insight.

“Ideological solidarity is a brittle glue, and reliance on it for intraparty cohesion is a sign of a weak party machine, not a strong one.” “Amateurs and single-issue activists have their uses, but without hacks the machine fails, which is why it puts such a premium on people who pay their dues and work their way up the ranks.” Richard Pildes: “broader structural changes, including legal ones, have disarmed party leaders of the tools they previously had used to unify their members around deals that were thought to be in the best interest of the party as a whole.”

Rauch points out that transparency makes certain kinds of politics and political agreements harder, because they look bad even though their results are good on balance. “In full public view, complicated deal-building is hard to do, indeed usually impossible; therefore machines tend to prefer privacy.” “[T]here is a cost to exposing the inner workings of politics to 24-7 public scrutiny”. “[T]he opposite of transparency in politics is not corruption, it is privacy”.

Rauch is good at demonstrating how successive campaign finance laws have just pushed money around and changed political incentive structures, without actually reducing the total amount of money in US politics.


The first major problem with the book is that Rauch doesn’t give much evidence for his case. It may be that machine politics produced better results… or it may not. Where’s the evidence? It is true that votes in legislative committees will have different outcomes depending on whether the way individual legislators voted is recorded or not. But it is not clear which is “better”. “If the machine gets its way, an independent political entrepreneur will have a difficult time influencing the system without paying her dues and going through channels.” Is this rent-seeking supposed to be obviously or inherently a good thing?

“They also act as gatekeepers to power, a function which today’s system of primary elections has weakened—at high cost to leadership in government. Imagine how dramatically the dynamics on Capitol Hill might change if state Republican parties gave Speaker Boehner the power to strike House candidates from the primary ballot: overnight, the party would move dramatically toward parliamentary-style discipline.” Again, it is not obvious that this would be an improvement — it needs to be argued.


The second, more serious problem with the book is that Rauch apparently never quite realises he is arguing against democracy. Early in the book he says that “transactional politics… is the essential work of governing and that government, and thus democracy, won’t work if leaders can’t make deals and make them stick.” Here he is equating good governance with democracy. It is true that government won’t work if politicians can’t make deals and make them stick, and that checks on politicians’ power make this harder. But taking power away from ordinary people, removing checks on politicians and enabling transactional politics, makes the system less democratic, while improving the quality of governance.

“Hence, a mayor successfully gaining sufficient influence over each of these institutions so as to make coherent policy could plausibly be seen not as a violation of democracy, but as a brilliant exercise of leadership so as to advance the interests of the poor as well as the public interest writ large.” Again, Rauch equates good governance in the public interest with democracy, yet they are not the same thing at all. “A mayor successfully gaining sufficient influence over each of these institutions so as to make coherent policy” is “a violation of democracy”, but (in this case) it is for the public interest. (We know from Arrow’s Impossibility Theorem that there is no guarantee that democratic government policy will be coherent.)

Indeed, the very title of the book is contradictory. “How Hacks, Machines, Big Money, and Back-Room Deals Can Strengthen American Democracy”. Replace “Democracy” with “Governance” and it makes sense.

Rauch quotes some people who do seem to realise the disconnect between good governance and democracy.

“Pildes argues for acknowledging a “tragic trade-off” between unmediated popular participation and government’s effective functioning. There may be no perfect balance between the two, but just accepting that more democracy is not always the answer would mark a sea change in American political discourse.”

Thad Williamson, University of Richmond: “To demand high standards of democratic deliberation… is to potentially slow, dampen, or even deny the ability of energetic leadership to play its mobilizing function and enact effective policies in a timely way.” (“The Tangled Relationship of Democracy, Leadership, and Justice in Urban America: A View from Richmond” (2014).)

Jason Grumet, “Bipartisan Policy Center”: “A closer look reveals that many of the reforms we’ve instituted to make government work more effectively—all of them pursued with the best of intentions—have, in practice, had the opposite effect. Through a series of bad assumptions and unintended consequences, we have weakened our government’s capacity to solve problems.” (“City of Rivals: Restoring the Glorious Mess of American Democracy” (2014).

Though, judging by the title of that last one, perhaps not.


Rauch views the purpose of politics as to reach compromises between various interest groups, which supposedly represent the whole of the population. This is a demotist view. By contrast, for neoreactionaries, the purpose of government is to run the country well, and politics is sand in the gears. For neoreactionaries, the residents of the country should be listened to like customers, but their demands for the divvying up of rents should be ignored.

Rauch has a story about the passage of some gay rights legislation in Utah. A compromise was reached between gay activists and Mormons. “Not everyone went away happy. Purists among libertarians, gay rights activists, and social conservatives all found reasons to grumble. But the negotiation and its outcome showed up the purists for what they were: outliers. And it built a consensus strong enough to withstand their vetoes, allowing implicit majority opinion to be heard.” “Implicit majority opinion”? What fiction is this?

For Rauch, an advantage of machine politics is that it causes “moderation”. “[M]odern analysts miss something important when they assume that moderation comes from moderates; often, moderation comes from machines that force politicians to compromise.” “Small donors tend to be polarized and polarizing.” “[A]s our campaign finance system has become more democratized, our politics has become more polarized.”

However, “moderation” is not an ideology advocated by specific people, but simply the contingent result of compromise. Moldbug: “Moderation is not an ideology. It is not an opinion. It is not a thought. It is an absence of thought.” “Moderation” is not much of an argument in favour of machine politics.


Rauch writes that “well-functioning governments rely not merely on formal legal mechanisms but also on informal political structures and intricate systems of incentives. No informal structures and incentives? No governance.” I’ve been thinking about this for a while.

When Rauch says that “the right amount of corruption is greater than zero”, he may be more right than he knows. Instead of just defining “corruption” as anything non-democratic, let us use Moldbug’s definition: “Corruption is any human action that is not what it appears to be.” Here Moldbug is making a moral assertion against informal power, property and relationships. Moldbug advocates formalism: the formalising of informal power, property and relationships.

But Jim points out that some situations cannot be formalised.
Moldbug advocates retiring useless government employees on full pay, to allow them to go off and do something useful — formalising their income stream. But in a democratic system this could not happen; voters would demand the abolition of the payments, and so useless bureaucrats would never agree to it, like Jim’s parasites: “The moment we come out, you will wrap us up in newspaper, and cast us onto the fire.” A private government would not need to formalise the income streams: it could just abolish them. Moldbug advocates formalism as a mechanism for getting from democratic government to private government. But it is hard to see circumstances in which it would happen; it is hard to see circumstances in which it would be the best course of action.

So if certain informal relationships are essential to good governance, and they can’t be formalised, then the right amount of “corruption” is greater than zero.

Neoreactionary research project: under what circumstances can informal arrangements be formalised without changing or destroying them?


I can’t resist quoting Rauch on the evils of tobacco: “Sometimes… the only thing wrong with smoke-filled rooms is the smoke.” Is smoke really so bad?


Available on Kindle.


More at Anomaly UK.