Creating Safe Spaces or Tone Policing?

How and when to moderate social media comments in a democratic campaign

Do platforms like Facebook and Twitter foster freedom of speech or inflame hatred, if they don’t take down incendiary comments? Should social media spaces be safe and inclusive, or is censorship a slippery slope? And which culture or perspective provides the right benchmark for what should or shouldn’t be allowed? Despite having teams of experienced policy professionals working on these questions daily, Big Tech still can’t get this right. So what hope for your own campaign?

The good news for campaigners is we don’t have to present ourselves as impartial arbiters of truth; we can set ground rules which feel right for our organisation and campaign alone.

Red lines 

Setting out a clear comments policy in your About section on your ‘red lines’ is a good starting point; any comments which threaten violence, or are racist, homophobic or in any way threatening to a protected group should be deleted, the perpetrator banned, blocked or hidden, and reported to the social media platform. (Hate speech, threats or incitement to violence are red lines for most social platforms too, thankfully).


Don’t get into arguments or even respond to hate-filled messages or threats; as my compatriot George Bernard Shaw says, “never wrestle with a pig. You both get dirty, and besides, the pig likes it”.

Facebook, Instagram and YouTube all offer ‘offensive content’ filters which can be switched on, and will automatically hide any comments using standard profanities (you don’t need to come up with a list of curse words to include, alas). YouTube’s ‘potentially inappropriate comment’ hold feature (which includes profanity as well as spam) is switched on by default, since June 2020.

Blocked word lists offer, as well as profanity blocking, the option to add other words which, though they might be inoffensive in other contexts, in a political campaign could become weaponised (like, for example, the word ‘snowflake’). Selectively adding additional words to these lists can enhance automated blocking and help reduce the amount of comments your social media team need to reactively review.  

Fuzzy lines

The question of what level of discourse you’ll allow or not allow in your social comments section is very subjective and something that you should decide as an organisation. Some organisations and campaigns might be very liberal in terms of freedom of speech, not wanting to censor any voices even if we disagree with them strongly, so as to foster healthy cross-spectrum debate. Other campaigners might feel they want their online spaces to remain civil, or family-friendly, or non-partisan or whatever it might be, and they have no qualms at all about shutting down anyone who crosses those lines.

ODV Digital is currently working with, a committedly nonpartisan campaign to improve democratic literacy and political engagement in Texas. We’re clear that the content we produce, both on site and on shared platforms, should speak across the political spectrum. But what of our newly-launched social comments sections? So far, we’ve seen hyper-partisan campaign slogans proliferate, not nuanced debate.

Restricting partisan comments or discourse is tricky because, let’s be honest, very few of us as individuals are actually politically neutral, even if our campaign or organisation is nonpartisan. We might believe firmly in the neutrality of the democratic process, but if we’re politically engaged, it’s very hard not to hold a partisan opinion one way or another. If we admit that about ourselves, should we expect it of others? Is policing a comments section to maintain non-partisanship fair, or even achievable? That said, it is legitimate to want to keep a digital space free of electioneering noise. So you might decide to have a policy which does not allow posts which simply post slogans but do not contribute to a debate.

Battle lines

An example might be: Texas2020 posts a video about Voting By Mail. A commenter might say “Trump is trying to stop us from being able to do this!” and another might reply, “Nonsense, the President is protecting us against electoral fraud!” We’ve decided that we would allow partisan comments like this, as they contribute to a healthy discussion of the subject. Where we nudge into the fuzzy lines would be if Commenter A added “This is why we MUST vote Democrat!” or Commenter B added “MAGA!”, as this then crosses the line into partisan electioneering.

Texas2020 has decided not to hide or block either A or B for either of the above; instead, we’ll be replying to remind commenters that this is a non-partisan, non-campaigning space for discussion of democratic processes (with a small d). If these comments persist, or if the comments did not include a meaningful response to our content and just became pure partisan electioneering, then we will hide those comments. If that same commenter persisted again, we would consider banning or blocking them (social media management software like Sprout can help identify ‘repeat offenders’ in content moderation).

Take the time to set out your ground rules, and make sure everyone in your organisation is in agreement. Then write out a clear content policy and include it (or a link to it) in your social accounts’ “about” sections.

Red lines: the ‘absolutely not, never’ content, including anything hate-filled or violent. Your own organisation’s red lines may go further than this; just be clear about where those lines are drawn.

  • Don’t respond to or engage in debates with people who cross these red lines. Delete the comments and ban or block the person.
  • Any words used in ‘red line’ contexts which are not automatically blocked by social platforms should be added to your accounts’ blocked word lists.

Fuzzy lines: areas for review and moderation. These might include hyper-partisanship, electioneering, the use of mild profanity, etc. Decide a policy for these fuzzy areas, which categorises when and how you should:

  • Respond to comments, reminding your audience of your campaign’s social content policies.
  • Hide individual comments
  • Ban or Block repeat offenders

Finally, be open and responsive to community feedback. A campaign lives and dies by the engagement of its public stakeholders. If people are not responding in ways with which you are comfortable, or if there is unanticipated thrash, perhaps your campaign needs to reconsider your target audiences or the messages you’re bringing them, or maybe you need to hear the voices of those who disagree with you.


Ultimately, however, we must do as Eleanor Roosevelt entreats:

“Do what you feel in your heart to be right – for you’ll be criticised anyway. You’ll be damned if you do, and damned if you don’t.”

The policy makers at the social media firms must surely agree.

Scroll to Top