bidding foster-mothers and nurses to suckle and bathe and wash the chidren, but in no wise to prattle or speak with them; for he would have learnt whether they would speak the Hebrew language (which had been the first), or Greek, or Latin, or Arabic, or perchance the tongue of their parents of whom they had been born.

Frederick’s experiment failed to demonstrate that Hebrew was the true prelapsarian language, as "without clappings of the hands, and gestures, and gladness of countenance, and blandishments" the babies promptly died. Psammetichus, the pharaoh who (so Herodotus reports) performed a similar experiment, had slightly better results; based on the similarity between the children’s babbling and becos, the Phrygian word for bread, Psammetichus was able to conclude that Phrygian was the oldest tongue. The experiments were probably doomed to failure; feral children generally never learn to communicate at an advanced level.

As recorded by the thirteenth century Franciscan Salimbene of Parma, the Holy Roman Emperor Frederick II engaged in a number of unusual experiments, including collecting orphaned babies and

bidding foster-mothers and nurses to suckle and bathe and wash the chidren, but in no wise to prattle or speak with them; for he would have learnt whether they would speak the Hebrew language (which had been the first), or Greek, or Latin, or Arabic, or perchance the tongue of their parents of whom they had been born.

Frederick’s experiment failed to demonstrate that Hebrew was the true prelapsarian language, as "without clappings of the hands, and gestures, and gladness of countenance, and blandishments" the babies promptly died. Psammetichus, the pharaoh who (so Herodotus reports) performed a similar experiment, had slightly better results; based on the similarity between the children’s babbling and becos, the Phrygian word for bread, Psammetichus was able to conclude that Phrygian was the oldest tongue. The experiments were probably doomed to failure; feral children generally never learn to communicate at an advanced level. Psycholinguists believe that there is a critical period in childhood language acquisition. One of the best known American feral children, Isabelle, was able to learn to speak normally, and Kaspar Hauser learned to speak reasonably well, but both seem to have had some human contact. One doesn’t need to be a fan of Steven Pinker to believe that children have a knack for languages and that it vanishes if it’s not developed. When two languages collide, the mishmash that springs up — think of Franglish — is a pidgin. When a pidgin becomes nativized, it is a creole (although some linguists think the relationship is more complex), and one of the key factors in the development of creole languages is the presence of children. Creoles can arise unexpectedly; a lack of experience on the part of the Managua’s first school for the deaf staff led children in Managua’s first school for the deaf to create a sort of sign language creole, Nicaraguan sign language, distinct from the world’s other sign languages.

Of course, like anything young children can do, natural languages tend to be sloppy and messy. Philosophers as early as the seventeenth century contemplated the creation of a new "philosophical language" that would allow them to transcend the unclarity of everyday speech. This proved even less successful than orthographic reform has over the years (after all, Noah Webster forced some of his ideas into American English where they remain today), and the effort to create a language based on a universal taxonomy of all possible objects and ideas never got much further than the efforts of Bishop John Wilkins, although the project did inspire work by Peter Roget, author of the noted thesaurus. Other artificial languages sprung up in the nineteenth and twentieth centuries: Interlingua, a sort of simplified Latin; Esperanto, an artificial Romance language; Ido, a revised Esperanto; Ceqli, a language which borrows grammatical elements from English and Chinese; Volapük, a Germanic language devised as the first planned intentional language; and dozens more. Many were invented for fictional society in the tradition of the work of noted philologist J.R.R. Tolkien; some of these, most famously Klingon, have attracted real-world students.

Bahasa Indonesia, an adapted Malay trade dialect, is the greatest success of any artifical language, with tens of millions of speakers throughout the huge (and linguistically far-flung) nation speaking the language. None of the languages designed to be international neutral tongues have had anywhere near that degree of success. Volapük and Esperanto flourished for a time, and Esperanto supporters occasionally asking the United Nations to adopt the language alongside, but neither language achieved widespread adoption. Nor, for that matter, did Solresol, an early attempt to create an international language based on musical notes and bearing a certain resemblence to both Wilkes’ philosophical language and the various whistle languages used in South America and the Canary Islands.

Of course, if the spirit of scientific inquiry is your goal, success doesn’t matter. The "women’s language" Láadan (link via MeFi) was created for a novel so that the author could make a point about the lingustic determinism elements of the Sapir-Whorf hypothesis. Loglan, the "logical language", was created for much the same reason:

Sapir and Whorf thought they had seen a pattern of relationships between the cultures of certain peoples and the structures of the individual languages with which those cultures were associated. That repeated pattern seemed to be containment. It was, they thought, as if human cultures were contained in their own languages, that each language set limits on the minds of its monolingual speakers so that each culture was constrained in its development by the very structures of the language in which it was expressed. Thus individual human cultures seemed not to develop in certain directions but to develop quite freely, even luxuriantly, in others…. [L]et us distinguish between two quite distinct kinds of Whorfian hypotheses: those that offer containment by domain-restriction as their explanation of all or some of the Whorfian phenomena, and those that offer facilitation or enablement…. What kind of hypothesis is Loglan good for testing? Well; both kinds. Loglan is both enabling and domain-enhancing.

The efforts may be Quixotic, but the intentions are good. So, too, are the intentions behind Lojban, a splinter language that was created when some Loglanists began to chafe at the control wielded by Loglan’s creator. There is no way to know if speakers of a language based on predicate calculus could have avoided messiness, but the Loglanists and Lojbanists settled it the most natural way for speakers of organic tongues: a lawsuit. The judgment was rendered in English.