In the history of analytic philosophy, there is much debate about whether a cryptomnesic experience counts as a memory. Bertrand Russell [1] thought that a real memory needed the bearer to have the sense of familiarity that it was indeed a memory. His thinking is certainly shared by Hume [2], who argued that memories, as opposed to imaginings, were particularly vivid to the person experiencing them. Martin and Deutscher [3] pointed out that the necessary conditions of a memory were not the phenomenal experience of it as a memory, but the correct causal connection to the perceptual event that precipitated it. Contemporary cognitive science echos Martin and Deutscher's observation. It recognizes a variety of memory systems, some of which may be active during cryptomnesia and some--such as conscious familiarity--may be passive, making any binary classification of Helen Keller's experience insufficient to capture the complex cognitive processes occurring in her mind at the time she wrote her story.
Philosophical issues aside, here is a great video of Helen and her teacher Anne Sullivan:
Thanks to
References:
[1] Russell, B. Analysis of Mind
[2] Hume
[3] Martin. C.B. & Deutscher M. (1966) Remembering. The Philosophical Review. 75(2). 161-196