You're in a bustling street market in Salvador, Brazil. You look up, and there she is. Your eyes meet, your heart skips a beat. It's love at first sight.
You are lucky enough to date, to marry and to have a child together. Unfortunately you happen to lose your job, and so you return to the UK to take up clerical work.
Fast-forward a few months to the autumn of 2010. You assume you can bring your wife and adorable one-year-old with you. There's one problem -- your wife grew up in the countryside in Maranhão, one of the poorest states in Brazil.
She lived on less than a dollar a day for most of her life. She has no formal education and cannot speak English. You're on a small salary. You already have to prove to the immigration authorities that you've got enough money to support and accommodate her, your child and yourself.
You struggle to scrape together the £1,000 or so needed for her English tuition and test fees. She studiously attempts to learn the language for eight months, but fails her test.
She retakes the exam several months later after yet more costly English classes, but fails again. It's been one and a half years and you've been living without your wife and child who still cannot join you. There is no end in sight.
This would be a perfectly ordinary example of how the coalition government's pre-English language tests for visas can be expected to impact upon those applying to join their British or settled spouses or partners.
The tests are officially justified in benign terms -- it's apparently all about "integration" and removing "cultural barriers".
Andrew Rosindell, the Conservative MP who appeared on The Politics Show over the weekend, was a little more upfront about the measures. Observing that the measures would lead to a 10 per cent reduction in applications, and therefore cut immigration numbers, he considered this to be a "good thing" and, indeed, one of two aims behind the measures. But however you look at these proposals they are open to serious objection.
Perhaps the ugliest aspect is the sheer discrimination and elitism that underpin it. The tests don't apply to spouses who cannot speak English from within the European Union.
For those to whom they do apply -- who happen to be drawn disproportionately from Asia, in particular south Asia -- it is very clear that the poor are most likely to suffer, as they are the least likely to be able to afford English lessons.
Given that family-based migration tends to be feminised, and that educational opportunities are gendered across the globe, it is highly likely that women will be disproportionately affected by all of this. Those with learning difficulties and mental health problems will also suffer.
The Netherlands recently introduced a similar scheme that has already been subject to international criticism by the United Nations Committee on the Elimination of All Forms of Racial Discrimination. The exemption there for "western states" was considered to have been discriminatory and inconsistent with international law.
But even from the point of view of efficacy in facilitating migrant integration, the measures are seriously lacking. First, how an earth can migrants who are prevented from coming to the country seriously be expected to integrate in the UK? The experience of the Netherlands shows any delay in family reunification/formation is likely to delay integration.
Further, as the proposals apply only to one small subsection of the migrant community in the UK, they are inherently limited in their reach, and therefore their ability to enhance integration is seriously compromised.
Moreover, these individuals are already subject to language and cultural testing at a more exacting level shortly after their arrival in the UK. Spousal testing in fact takes place within 24 months after arrival.
If anything, these measures will in the long term hinder integration. They will only reinforce erroneous stereotypes held by some sections of the host community that migrants on the whole don't wish to learn the English language or be part of the UK. From the point of view of migrants, the discrimination and hardship that the measures will subject them to will delay the development of the sense of membership of a society that is a precondition for integration.
The experience of the Joint Council for the Welfare of Immigrants is that most migrants do in fact learn the language and actually want the opportunity to do this but oversubscription for classes, and the costs of learning tend act as significant barriers.
A far more effective and proportionate way of addressing any deficiencies in the English-language skills of migrants would be to remove restrictions that prevent many migrants who arrive in the UK from learning the language at an earlier stage. This could be done through charging fees at more affordable home prices, rather than overseas rates, which tend to be prohibitive.
Further, the £50m the government proposes to spend on implementing this scheme could be better invested in expanding already oversubscribed English language classes.
This would give all immigrants -- and all others with less sophisticated written and spoken English skills to boot -- the opportunity to improve.
Adopting such an approach to language learning, integration and immigration policy would also reflect international law and the principles that underpin it.
Indeed, international consensus on the centrality of family unity to human dignity crystallised as early as 1948 with the UN Declaration of Human Rights. Dog-whistle politics offers no justification to depart from these obligations or the long-cherished principles on which they are based.