That the backstop is "anti-democratic" has become a familiar refrain of its opponents. In this way, for some it stands as a handy validation for Brexit itself.
If the backstop means taking rules from Brussels with "no say over them", then it is obviously anathema to the quest to reclaim national sovereignty.
But ensuring that people in Northern Ireland have an influence over the operation of the backstop is less a matter of democracy than one of legitimacy.
Legitimacy is derived from the quality of relationship that exists between people and political systems. Brexit means a fundamental change in the political system of the UK. That system is being readjusted to exclude the institutions and processes of the EU.
This has particular implications for Northern Ireland, which will continue to have close economic, administrative and legal ties with Ireland, an EU member state.
These ties can become thinner and looser, but they cannot be completely cut. In this regard there can be no clean break of the UK from the EU.
The Protocol on Ireland/Northern Ireland in the withdrawal agreement (from which the backstop arises) is an effort to create a legally sure environment for continuing many of these ties.
This has potential implications for laws, policies and other binding decisions that affect Northern Ireland. As such, it gets right to the heart of the matter of legitimacy.
The political system in Northern Ireland will inevitably change after Brexit. It is useful, then, to consider what type of legitimacy is preferable and viable in this context.
There is a simple typology that can help us assess the quality and nature of legitimacy of a political system.
First, there is 'input' legitimacy: based on participation in decision-making, ie who is involved. Second, there is 'output' legitimacy: based on performance, ie what is the outcome of decisions made. Third, there is 'throughput' legitimacy: based on process, ie how are the decisions made.
There are trade-offs between these different types of legitimacy.
The governance of Northern Ireland since the 1998 Good Friday (Belfast) Agreement has been based on input legitimacy - on the assumption that there are two communities in Northern Ireland who identify with and aspire to two different states.
This is reflected in the three strands of the Agreement (British-Irish, north/south, unionist/nationalist) and in some of the unique features of the devolved decision-making, such as the provisions for parallel consent between unionists and nationalists.
Similar assumptions have been brought through into some people's view of the backstop. Most prevalent of these is the assertion that there needs to be consent of both unionists and nationalists in the operation of the backstop.
But if unionists and nationalists could separately have a veto over the operation of the backstop, would that be adequate assurance of its legitimacy? The solution for the governance of post-Brexit Northern Ireland will have to go much further than straightforward input legitimacy.
This is, in part, because - put bluntly - input legitimacy vis-a-vis applicable EU acts is not on offer for non-members. No MEPs, no seat on the Council, no Commissioner.
What about output legitimacy? After Brexit, we are in an entirely new zone. The only way in which people could assess whether the system is performing well would be to compare it against what went before (a false comparison), or else the alternative (which would be untried and thus unquantifiable).
So, this leaves throughput legitimacy. There are two dimensions to the process for throughput legitimacy that need to be covered. We can summarise them as 'uploading' and 'downloading'.
Uploading means getting voices and views from Northern Ireland heard on the operation of the backstop.
Northern Ireland needs to be able to influence London and Brussels. Within the UK, this will require not only proper mechanisms in inter-parliamentary bodies (eg a joint committee between Stormont and Westminster), but also effective input into the formulation of Government policy.
This could be through a reformed Joint Ministerial Committee that connects ministers from London to those in the devolved regions and nations.
In Brussels, representation from Northern Ireland on the UK-EU bodies provided for in the withdrawal agreement would be a strong start; for example, the First and Deputy First Minister being allowed to attend relevant parts of the UK-EU Joint Committee.
Input from the 1998 Agreement institutions and bodies could play an important role here. There could also be space for social partnership forums, or civil platforms, as happens currently in the EU for the eastern partnership countries preparing for accession.
A cross-border dimension to these conversations is also needed - potentially east/west as well as north/south, given the possible scope of the backstop.
Then there is downloading, in other words ensuring legitimacy in managing the application of EU rules in Northern Ireland. At the very least, there will be need to be mechanisms in place for the effective scrutiny of legislation and running impact assessments.
Whether the heavy weight of this occurs in London or Belfast depends on whether Stormont is functioning and what additional responsibilities the UK Government wishes to devolve.
Once we recognise that alignment is needed to minimise friction on movement across the border, the question should be how best to secure legitimacy in the system by which this occurs - and how to do so in a way that works best for Northern Ireland. This will require action from London, as well as accommodation from Brussels.
Katy Hayward is a Reader in Sociology and David Phinnemore is Dean of Education and Professor of European Politics at Queen's University Belfast