◎ OireachtasDB

← Leo Varadkar

Rhetoric and votes

Leo Varadkar

Experimental measure. Read the method before quoting a number.
This compares the position a member expressed in a debate with the way they voted in the division that debate produced. It is not a measure of honesty. A member who argues against a Bill and then votes against it scores as aligned; so does one who never departs from a party line they also argue for. Voting against your own stated view under a party whip is ordinary parliamentary behaviour, which is why the whip breakdown below sits next to the headline rather than underneath it. Full method and validation →
82%
aligned on 11 scorable votes · 95% CI 52–95%
100%
voted with their own party's majority · 441 votes

Why the denominator is small

The measure only exists where this member actually spoke in the debate their vote belongs to, and where that vote can be tied to a specific question. Both cuts are steep.

Votes cast443
… spoke in that debate section44
… vote tied to a specific question36
… position stated clearly enough to score11

The third row is the sharpest cut. A committee-stage sitting can hold fifteen divisions while its speeches discuss forty-two numbered amendments, and the record does not say which amendment each division was. Those votes are counted here and deliberately left unscored rather than guessed.

Alignment against the whip

The same scorable votes, split by whether the member went with their party. The top-right cell is the one worth reading: a stated position reversed while following the party line.

Voted with own partyVoted against own party
Matched their words 9 0
Contradicted their words 2 0

Not scored: 8 where the vote could not be tied to a question, 0 where no clear position was stated in the debate.

Where the words and the vote diverge

Every case the pipeline flagged, with the contribution it read and the vote as recorded. Judge them yourself — that is the point of showing them rather than only the rate.

2023-09-26 · An tOrd Gnó - Order of Business · Question put: · Carried

Read as oppose (cue, confidence 0.90) · voted TA · with their party

All four of the companies the Minister, Deputy Ryan, and I met have announced reductions in their prices. Since then, some other companies have announced competitive prices as well. However, the message from us to them was very clear. We do not think their prices are coming down enough. We want to see further reductions in the next billing cycle in the new year. They pointed out to us that there is not a direct link …

Full contribution → · The division →

2021-11-02 · An tOrd Gnó - Order of Business · Question put: "That the proposal for dealing with this week's business be agreed to." · Carried

Read as oppose (cue, confidence 0.85) · voted TA · with their party

The power stations at Lanesborough and Shannonbridge were not closed as a consequence of a Government decision. It is important to emphasise that. I know the Deputy is aware of it already. These were decisions by An Bord Pleanála and the courts. What the ESB wanted to do and what the Government wanted it to do was to convert those stations to biomass, such that they would use a mix of peat and biomass and then biomas…

Full contribution → · The division →

Tone

Valence runs −1 (negative) to +1 (positive). This is a separate measure from alignment above and says nothing about how anyone voted — "the Minister has been asleep at the wheel" is strongly negative and carries no position on the question.

ModelSpeeches scored Mean valenceCorpus mean
vader4,899 +0.482 +0.423

Which policies move them

Mean tone by topic, most negative first. Topics come from an LDA model fitted over debate contributions; the terms are the model's, the labels are hand-assigned.

TopicSpeechesMean valence
Health scandals, inquiries and bereaved families hospital · families · family · cancer · children · did 16 -0.103
Housing, rent and homelessness housing · homes · local · accommodation · rent · affordable 24 +0.324
War, Ukraine, Israel and human rights international · ireland · ukraine · israel · war · irish 31 +0.330
Legislation, courts and legal process act · legislation · shall · court · person · information 55 +0.365
Health services, HSE and disability health · services · care · service · hse · mental 41 +0.419
Employment, welfare schemes and business supports scheme · businesses · payment · sector · employment · insurance 107 +0.436
Media, banking and State companies ireland · irish · media · industry · companies · post 10 +0.442
Covid-19 and public health emergency covid · health · public · pandemic · measures · safety 75 +0.447
Justice, policing and defence garda · defence · justice · violence · criminal · gardaí 24 +0.460
Cost of living, tax and budget cost · tax · budget · energy · costs · increase 52 +0.516
Procedural and rhetorical filler (low-content) important · like · good · debate · did · does 179 +0.560
Children, families and young people children · women · young · child · family · families 5 +0.566
Climate, agriculture, energy and water water · climate · farmers · energy · food · agriculture 10 +0.643
EU and international relations ireland · european · irish · union · europe · countries 37 +0.651
Party-political exchange (party names dominate) workers · rights · right · sinn · féin · irish 23 +0.663
Education and schools school · education · schools · children · students · special 19 +0.674
Local government, roads and regional funding county · cork · discuss · west · road · local 14 +0.726
Departmental programmes and funding (diffuse) department · national · funding · new · important · programme 54 +0.759
Transport, Dublin and rural development local · transport · dublin · city · public · community 19 +0.842

Departures from their own baseline

Contributions whose tone sits more than two standard deviations from this member's own average. Baselines are per member on purpose — an absolute threshold would just rank people by temperament.

2022-05-18 · Sick Leave Bill 2022: Report and Final Stages · z = -2.8 · Legislation, courts and legal process
I move amendment No. 30: In page 11, line 22, to delete “20 weeks’” and substitute “4 weeks’”. These amendments amend the maximum penalty the Workplace Relations Commission or Labour Court may award if an employer breaches obligations under the Bill. That now …

Full contribution →

2023-07-04 · Ceisteanna ó Cheannairí - Leaders' Questions · z = -2.8
We always consider the proposals that are put to us by Opposition parties, NGOs and others working in the area of homelessness, as well as, of course, those who build homes – the home builders. For example, we accepted the proposal to introduce a rent credit a…

Full contribution →

2022-12-13 · International Agreements · z = -2.8
The convention is a legal instrument that recognises the right of everyone to a world of work free from violence and harassment, including gender-based violence and harassment as well. It acknowledges that gender-based violence and harassment disproportionatel…

Full contribution →

2022-10-06 · Ceisteanna ar Pholasaí nó ar Reachtaíocht - Questions on Policy or Legislation · z = -2.8
Thankfully, evictions in this country are relatively uncommon. They need to be done through courts and judges look very favourably on very difficult situations in that regard. Notices to quit are a bit different but if somebody is given a notice to quit, they …

Full contribution →

2023-11-21 · Cabinet Committees · z = -2.8
Deputy Barry raised the issue of gender-based violence, violence against women and femicide. As I have said previously, I believe we face an epidemic of violence women not just in Ireland, but across the world. As the Deputy said, there have been approximately…

Full contribution →

2023-11-15 · Ceisteanna ó Cheannairí - Leaders' Questions · z = -2.8
Those are the two reasons the Government thinks Sinn Féin's proposal would be of no use and would be unbalanced. What we are going to do is to provide an additional €3 million to the International Criminal Court. This will be of practical assistance and help i…

Full contribution →

2020-06-04 · Covid-19 (Taoiseach): Statements · z = -2.8 · Party-political exchange (party names dominate)
I answered a similar question on direct provision earlier. On the events we have seen take place in the United States, as I stated, we have witnessed a real absence of moral leadership from the top in the United States. We should have had words of unity, comfo…

Full contribution →

2023-03-22 · Cabinet Committees · z = -2.8
I do not think any modelling has been done by the Opposition. We had an indication of the RTB figures but not the exact figures. They indicated two things. One was an uptick in notices to quit driven by the eviction ban. Landlords got the message, in part as a…

Full contribution →

2023-02-08 · Ceisteanna ó Cheannairí - Leaders' Questions · z = -2.8
None of us in this House wants to see the issue of race or migration become centre stage in our politics, most of all somebody like me, given my colour and my family background and the fact I am biracial. I ask the Deputy to consider his approach to this, and …

Full contribution →

2024-09-18 · An tOrd Gnó - Order of Business · z = -2.8 · War, Ukraine, Israel and human rights
I thank the Ceann Comhairle. I raise with the Taoiseach the appalling catastrophe unfolding in Sudan. As the House may know, civil war has raged there for approximately two years, leaving 150,000 dead. Some 2 million refugees have left Sudan already, with mill…

Full contribution →

2021-11-25 · Ceisteanna ó Cheannairí - Leaders' Questions · z = -2.8
I thank the Deputy for raising the important issue of child protection. I realise that we will debate this in the House later and that we will continue to discuss it in the weeks and months ahead. I agree that we must do more to help every child to get the sta…

Full contribution →

2022-05-18 · Sick Leave Bill 2022: Report and Final Stages · z = -2.8 · Employment, welfare schemes and business supports
These things amount to a judgment call. My original view was the same as the Deputies in that it should be 20 weeks' remuneration but having taken advice, in particular legal advice, and heard the views of many others, we had to try to make sure the penalty wa…

Full contribution →

The whip, across the whole corpus

Alignment by the party a member sat for on the day, with 95% Wilson intervals. This is the test the measure could have failed.

PartyAligned95% CI n
PBP 82.5% 68–91% 40
LAB 78.2% 69–85% 101
IND 76.5% 72–81% 327
SF 75.8% 69–81% 198
SD 71.4% 58–82% 49
FG 69.1% 62–76% 162
FF 67.9% 60–75% 165
GP 56.8% 41–71% 37

Members of government parties align their stated position with their vote 9.4 points less often than Independents — bootstrap 95% CI [+2.7, +16.0], which excludes zero. Opposition parties sit 9.5 points above government too, CI [+3.1, +15.9]. Independents and opposition members are indistinguishable from each other (−0.0 points, CI [−6.3, +6.2]). So the gap does not come from having no party; it comes from being in government, which is what a whip effect should look like. Read this as the measure behaving sensibly, not as a finding about any individual.

How the scorer scores

Two automated methods were built and both failed. What survives is the rule-based layer, which is why the coverage above is narrow.

MethodAccuracy95% CI n
nli-vs-mover-anchor:deberta 35.9% 29–42% 234
nli-vs-mover-anchor:distilroberta 11.5% 8–16% 234

Zero-shot entailment was asked to judge contributions whose true position is known — the member moved the amendment, so the answer is support by construction — and got it right 11.5% of the time, returning "unclear" or "oppose" for nearly everything. A classifier trained on the rule-labelled cases over the corpus embeddings reached 71.1% against a 65.3% majority-class baseline: better than chance, not better enough. Both are excluded from every figure on this page. Only the rule-based cascade, which is exact where it fires and silent where it does not, is used.

Method

Position is read from the member's own contributions in the debate section the division belongs to, by a three-step cascade: an explicit move of the question ("I move amendment No. 5", true by construction); declarative stance phrases ordered so negations are tested first, because "I cannot support" and "I do not oppose" both defeat a naive keyword match; and, where neither fires, a zero-shot entailment model judging the contribution against the question put. Tone uses a separate model and is reported separately. Agreement against hand-labelled cases, with bootstrap intervals, is on the evidence page; no score here should be read without it.